Is acceleration absolute and velocity not?

[frb]

is it, cause velocity isn't.

 

[dextercioby]

is it, cause velocity isn't.

Since coordinates,velocity and acceleration are ALL OF THEM CINEMATIC QUANTITIES,they need to be defined wrt a reference system.Therefore their character is RELATIVE AND NOT ABSOLUTE.

Daniel.

 

[kirovman]

is it, cause velocity isn't.

No, because if two objects accelerate towards each other you get the sum of their acceleration.

Same with forces, since F=ma.

For example ficticious forces centrifugal force, and centrifugal acceleration, are not considered in all reference frames.

 

[Andrew Mason]

It depends on what you mean by absolute. It is not invariant under relativity, but it is a quantity which all observers in all reference frames would observe and be able to measure.

If Einstein's principle of equivalence is correct, since acceleration and gravitation are equivalent, acceleration must be observed in all reference frames and observers in all reference frames must agree.

Ignoring relativistic effects, observers in all inertial frames would agree on the measure of acceleration, even though they are traveling at different speeds. If they are traveling at relativistic speeds, they would have to apply the Lorentz transformations.

AM

 

[Andrew Mason]

Since coordinates,velocity and acceleration are ALL OF THEM CINEMATIC QUANTITIES,they need to be defined wrt a reference system.

It depends on what movie you are watching?

AM

 

[Gokul43201]

I agree with AM.

You CAN determine acceleration absolutely, at least at low velocities. It is always possible to know if a frame is accelerating , using a simple pendulum or glass of water, or some such thing.

 

[ramollari]

Let's see it from a Newtonian perspective (low speeds). There are two inertial frames of reference that observe a moving body. According to frame 1 it is v1 and according to frame 2 it will be v2. So velocities are relative. If frame 2 is moving at a speed v' relative to frame 1, then according to the classic addition of relative velocities, v2 = v1 + v'

then,

a2 = dv2/dt = d(v1 + v')/dt = dv1/dt + 0 = a1,

since v' is constant (inertial frames of reference).

If velocities are measured in 3 dimensions, just split to their components, and apply the same method. You will get the same x, y, and z component for acceleration.

 

[marlon]

Acceleration is absolute in every theory before General Relativity. So, Newtonian physics (here, also speed is absolute), Special relativity (no absolute speeds, only absolute acceleration)

It is in General Relativity that acceleration is no longer absolute because every reference frame is equivalent.


regards
marlon

 

[Chronos]

Hmmm, where to start. The difference between general and special relativity is general theory treats all reference frames, including rotating and accelerating frames as equal. Special relativity treats accelerating frames differently. In SR, acceleration is an absolute and velocity is relative. In GR, all motion is relative. GR requires curved spacetime to make this work. This idea is not the least bit exotic. Solving classical mechanics problem in curved coordinate systems [e.g., polar coordinates] is a common and fully accepted practice. It makes sense and is much easier to use to analyze and make concrete predictions.

I should point out that if you adopt the proper coordinate system in SR, it produces the same result as GR. GR restricts your choice of coordinate systems, SR does not. That leads to many perceived paradoxes. It is also explains why Einstein took 10 years to publish his work on GR. The concept was too vitally important for Einstein to present anything other than an air-tight case. He has done a pretty amazing job so far... 90 years later and no one has yet punched an irrefutable hole in that theoretical edifice.

 

[marcus]

is acceleration absolute?

is it, cause velocity isn't.

what about rotation?

can an observer decide whether his or her frame of reference is rotating or not? can different observers agree?

 

[Rybo]

Rotation is constant/absolute accelearation is not.

what about rotation?
can an observer decide whether his or her frame of reference is rotating or not? can different observers agree?

Marcus, yes to both questions, provided they have the skill and observational equipment to calulate the realtionship of two or more frames outside of their immediate/local frame of refereence.


All things are in motion(energetic) ergo all things are either accelerating(out) or decelerating(in) to one or more other things in Universe.

Rybo

 

[jdavel]

marlon,

You said,

"Acceleration is absolute in every theory before General Relativity. So, Newtonian physics (here, also speed is absolute), Special relativity (no absolute speeds, only absolute acceleration)"

But unlike Newtonian physics, according to SR the magnitude of acceleration is different in different inertial frames.

 

[marlon]

But unlike Newtonian physics, according to SR the magnitude of acceleration is different in different inertial frames.

Well, the mere fact that you are talking about different inertial frames proves the fact that you need an absolute acceleration because how do you distinguish between an ordinary frame of reference and an inertial frame of reference.

Given the fact that all physical laws must be the same in each inertial frame i don't think i get your point. how do you mean acceleration is different?All physical laws (you know ma =...) are the same so ?

But nevertheless in SR acceleration is ABSOLUTE because of the very reason i gave above


regards
marlon

 

[ramollari]

Why is angular velocity absolute and linear velocity relative? Isn't that a paradox? What makes a frame of reference inertial in terms of rotation is really a mistery to me.

 

[Andrew Mason]

marlon,

You said,

"Acceleration is absolute in every theory before General Relativity. So, Newtonian physics (here, also speed is absolute), Special relativity (no absolute speeds, only absolute acceleration)"

But unlike Newtonian physics, according to SR the magnitude of acceleration is different in different inertial frames.

Actually it is not acceleration that is absolute. It is the rate of change of momentum. In SR, dp/dt is invariant and absolute.

AM

 

[jdavel]

marlon,

Maybe use of the word "absolute" is causing our disagreement. Can we change to "invariant"? That's very clearly defined in the context of any relativity theory.

For example invariants in Galilean relativity include acceleration, force, mass, time intervals between events, etc. But in general, not velocity, momentum, kintetic energy, distance between events etc.

Invariants in SR include the speed of light (of course!), the interval between two events etc. And there are special cases where dv/dt and dp/dt are invariant, but in general they aren't.

 

[reilly]

There's no way acceleration is anything but relative. It's the second deriv. of a distance, or coord, which necessarily refers to two points, and that's the source of the relative nature of spatial coordinates, and functions thereof.

Regards,
Reilly Atkinson

 

[salazar18]

need to brush up on physics

it seems that this dialogue is lacking in depth of knowledge of momentum, angular momentum, time-space curvature, general relativity, and cyclical energy fluctuations

 

[Andrew Mason]

it seems that this dialogue is lacking in depth of knowledge of momentum, angular momentum, time-space curvature, general relativity, and cyclical energy fluctuations

Then perhaps you would care to enlighten us. Why is/is not dp/dt at least absolute and invariant?

AM

 

[Andrew Mason]

There's no way acceleration is anything but relative. It's the second deriv. of a distance, or coord, which necessarily refers to two points, and that's the source of the relative nature of spatial coordinates, and functions thereof.

I don't think that can be right. Although position is dependent upon a co-ordinate system, change in position and rate of change of position is not (at least to inertial observers where relative v<<c). Mass is not dependent upon the frame of reference either (if v<<c). To any inertial observer, (even those moving at speeds close to c) dp/dt is invariant and absolute under Newtonian/Galilean relativity and under relativity.

AM

 

[jdavel]

Andrew,

"To any inertial observer, (even those moving at speeds close to c) dp/dt is invariant and absolute under Newtonian/Galilean relativity and under relativity."

That's not true! In general, a Lorentz transformation will change both the direction and magnitude of force. In fact force and the acceleration it causes can end up in mutually different directions after applying an LT!

Nature paid a huge price for insisting on there being one speed that would be the same in all reference frames! The price was that almost nothing else could be.

 

[reilly]

Fred's watching me go down in a free falling elevator -- Albert Einstein Model # 32g.
Unless Fred is deranged, he will say that I'm moving down with acceleration g, and he's not.
Of course, I say something different. Fred's moving up with g, and I'm not accelerating.
That's about as relative as it gets.
regards,
Reilly Atkinson

 

[Andrew Mason]

That's not true! In general, a Lorentz transformation will change both the direction and magnitude of force. In fact force and the acceleration it causes can end up in mutually different directions after applying an LT!

I don't follow you there. Force is defined as $\vec F = d\vec p/dt$. So it has to be in the direction of the momentum change, which means it has to be in the direction of the acceleration.

Nature paid a huge price for insisting on there being one speed that would be the same in all reference frames! The price was that almost nothing else could be.

If dp/dt is not invariant under relativity then momentum cannot be conserved. Neither can energy because energy is related to momentum by:

$$E^2 - (pc)^2 = (mc^2)^2$$

If you can explain why it is not with an example, it might help. But right now, I don't see it.

AM

 

[Andrew Mason]

Fred's watching me go down in a free falling elevator -- Albert Einstein Model # 32g.
Unless Fred is deranged, he will say that I'm moving down with acceleration g, and he's not.
Of course, I say something different. Fred's moving up with g, and I'm not accelerating.
That's about as relative as it gets.
regards,
Reilly Atkinson

I thought we were talking about inertial reference frames. Acceleration is relative under Newtonian mechanics if you use non-inertial frames of reference.

AM

 

[kawikdx225]

Fred's watching me go down in a free falling elevator -- Albert Einstein Model # 32g.
Unless Fred is deranged, he will say that I'm moving down with acceleration g, and he's not.
Of course, I say something different. Fred's moving up with g, and I'm not accelerating.
That's about as relative as it gets.
regards,
Reilly Atkinson

An accelerometer mounted to you and Fred will determine which one of you is accelerating. You will measure that you are accelerating and Fred isn't.

A spaceship in deep space can also measure absolute acceleration using an accelerometer, acceleration can be relative but doesn't have to be.

A rotating object is also accelerating absolutly, just spin real fast in a circle and see what your arms tend to do.

 

[jdavel]

Andrew,

I don't follow you there. Force is defined as $\vec F = d\vec p/dt$. So it has to be in the direction of the momentum change, which means it has to be in the direction of the acceleration.

It seems that way, but it doesn't because the mass changes as well as the velocity.

Suppose an object is moving in the x direction at .99c. Now apply a force to it with components Fx = 10 and Fy = 1. That's almost along the x direction. But look at the resulting acceleration. The Fx = 10 can barely cause any acceleration because the object is already almost going at c. But the Fy = 1 can increase the y component of velocity to 0.1c without the resultant velocity exceeding c. The force was almost all in the x direction, but the acceleration is almost all in the y direction!

 

[Andrew Mason]

Andrew,



It seems that way, but it doesn't because the mass changes as well as the velocity.

Suppose an object is moving in the x direction at .99c. Now apply a force to it with components Fx = 10 and Fy = 1. That's almost along the x direction. But look at the resulting acceleration. The Fx = 10 can barely cause any acceleration because the object is already almost going at c. But the Fy = 1 can increase the y component of velocity to 0.1c without the resultant velocity exceeding c. The force was almost all in the x direction, but the acceleration is almost all in the y direction!

The path of motion changes, but does the acceleration?
Your example applies equally to Newtonian mechanics as well. If I throw a ball sideways off a moving train it moves at 90 degrees to me but at a forward angle relative to the earth. If I make it a rocket, and give it acceleration, it accelerates straight out at 90 degrees but to an observer on the ground it moves in a curve (parabola). But both would agree on the direction and magntitude of the acceleration: it is a vector pointing perpendicular to the train.

In the relativistic case, I think inertial observers in both frames would agree on the direction and rate of change of momentum (ie the magnitude of the force and direction). They would just disagree on whether it was due to additional velocity or additional mass .

AM

 

[jdavel]

AM

"Your example applies equally to Newtonian mechanics as well."

No it doesn't; you misunderstood my example.Try this.

Suppose you're in a frame of reference where a particle is moving slowly (v<<c). If you apply a force to the particle its momentum will change at the rate given by dp/dt = d(mv)/dt = m(dv/dt). So dp/dt = ma. That's a vector equation and since dp/dt and a are the only two vectors, they have to be pointing in the same direction.

But now if you look at the same thing in a frame of reference where the particle is moving at .99c, the force will cause the momentum to change at the rate given by dp/dt = d(mv)/dt = m(dv/dt) + v(dm/dt). So dp/dt = ma + v(dm/dt). Again, that's a vector equation, but this time there are three vectors, dp/dt, a and v. So unless v is in the same direction as a, dp/dt will point somewhere in between v and a, not in the same direction as either one.

By the way this all works out in a very neat equation (as is so often the case with special relativity!) which says:

Fx/Fy = gamma^2*ax/ay

where Fx and Fy are the components of force and ax and ay the components of resulting acceleration. In frames where the particle's speed v<<c, gamma^2 is nearly 1, and you get the Newtonian result (force and acceleration are codirectional). But for speeds close to c, you don't.

 

[Andrew Mason]

AM

"Your example applies equally to Newtonian mechanics as well."

No it doesn't; you misunderstood my example.Try this.

Suppose you're in a frame of reference where a particle is moving slowly (v<<c). If you apply a force to the particle its momentum will change at the rate given by dp/dt = d(mv)/dt = m(dv/dt). So dp/dt = ma. That's a vector equation and since dp/dt and a are the only two vectors, they have to be pointing in the same direction.

But now if you look at the same thing in a frame of reference where the particle is moving at .99c, the force will cause the momentum to change at the rate given by dp/dt = d(mv)/dt = m(dv/dt) + v(dm/dt). So dp/dt = ma + v(dm/dt). Again, that's a vector equation, but this time there are three vectors, dp/dt, a and v. So unless v is in the same direction as a, dp/dt will point somewhere in between v and a, not in the same direction as either one.

Good point. The direction is not absolute - I retract my earlier statement about direction! But I think |dp/dt| is still the same for all observers. Try working it out and see.

AM

 

[jdavel]

AM,

"The direction is not absolute - I retract my earlier statement about direction! But I think |dp/dt| is still the same for all observers."

Nope! In general, the magnitude isn't invariant either. There's one special case where both the magnitude and direction of force are invariant to a LT. It's where the force and the initial velocity of the particle it acts on are parallel to the direction of relative motion of the two frames. Otherwise everything gets bent and squished.

But look on the bright side; if what you said were true, you couldn't hang stuff on your refrigerator! :-)

 

[prasanth.s]

Am new here
Lets limit our discussion to Classical Mechanics.

Any of the 'accelero-detectors' mentioned in a post above(simple pendulum, etc.) will work inside a spherical shell whirled around in a circle using a string(or inside a bus taking a turn). But will they work inside a satellite going around the earth?
Ans: No
Reason: Because it is in free fall
Natural follow-up question: So what? Why is that the reason?

This raises questions about which forces can be felt or detected(alternatively which accelerations are detectable)

My claim is that if you have a force acting on every point(mass) on a body producing same acceleration on each(like gravitational force except that g is constant throughout the body) then there is no way for the body to detect the force. It can only figure out if it is accelerating or not by observing its surroundings for RELATIVE accelerations.

And if there is type of force (theoretically-there may not be such a force falling into the list of four basic forces in nature) that can never be detected it means that ACCELERATION IS RELATIVE

 

[Mueiz]

is acceleration absolute..is it, cause velocity isn't.

In gravitational field it is absolute and equal zero in free-falling frames.
In the absence of gravitational field it is relative.

 

[Dale]

In gravitational field it is absolute and equal zero in free-falling frames.
In the absence of gravitational field it is relative.

But according to you there is always a gravitational field and therefore it is always absolute. Stop posting nonsense that even you don't believe has any physical meaning.

 

[Dale]

Am new here ... ACCELERATION IS RELATIVE

Hi prasanth.s, welcome to PF. Usually rather than responding to a post which is 6 years old it is better to start a new thread. That said, there are two kinds of acceleration that we can discuss, one is called "proper acceleration" and it is absolute, the other is called "coordinate acceleration" and it is relative. If you would like to discuss the difference I recommend a new thread.

 

[K^2]

what about rotation?

can an observer decide whether his or her frame of reference is rotating or not? can different observers agree?

You can set up a gravitational field that will make any local experiment confirm rotation without there actually being any rotation. Frame dragging can even simulate Coriolis Effect.