PDA

View Full Version : Now about them there gravity waves ...

David Simmons
08-12-2003, 04:49 PM
When the earth is between the sun and the moon, the em waves of light are blocked and the moon is dark. When x-rays pass through matter they are differentially absorbed. Gamma rays knock out pieces of atoms. All of these em waves interact with matter.

When the earth passes between the sun and the moon the effect of the sun's gravity on the moon, and vice versa, is undisturbed. Gravity waves must not interact with matter. At least in that case.

So what do they do? The best I can figure is that they distort space in the vicinity (well, to infinite distances) around objects with mass in such a way that curved coordinates are needed to describe it. But once a pattern of "bent" space is established around the mass, the effect stands still in space and time. Or does it?

Mayday!

MC Master of Ceremonies
08-12-2003, 06:36 PM
The Earth, the sun and the moon are in static gravitational fields so you shouldn't be able to observe any gravity waves between them, or in otherwords the graviton exchange is virtual just like the photons in a static electromagnetic field.

MikeS
08-12-2003, 07:57 PM
It's also important to remember that although Newtonian gravity is a good approximation to general relativity, there are properties of the Newtonian theory that don't carry over into GR. One of these properties is linearity, i.e. the statement that "force due to bodies A and B = force that would be exist if only A was present + force that would be exist if only B was present" is no longer true. It is true to a very good approximation, granted; but the non-linear effects are present (roughly speaking) at the same level of approximation as one would use for gravity waves, so if you're going to include one you'd best include the other.

08-12-2003, 07:59 PM
Additional question: the "super strings" folks (and others) expect to find gravitons--particles that carry the gravitic force. My understanding is that they would have the same speed-of-travel limit as other particles, i.e. 186m/sec. So how can a black hole have a gravitational field if the gravitons cannot escape?

Thank you.

JasonFin
08-12-2003, 08:07 PM
Additional question: the "super strings" folks (and others) expect to find gravitons--particles that carry the gravitic force. My understanding is that they would have the same speed-of-travel limit as other particles, i.e. 186m/sec. So how can a black hole have a gravitational field if the gravitons cannot escape?

Thank you.

David Simmons
08-12-2003, 08:48 PM
Originally posted by MC Master of Ceremonies
The Earth, the sun and the moon are in static gravitational fields so you shouldn't be able to observe any gravity waves between them, or in otherwords the graviton exchange is virtual just like the photons in a static electromagnetic field.

I'm going to take this as saying that my crude statement that once the pattern of "bent space" is established that pattern, or gravitational field, is stationary with respect to the mass that established it.

OK. So what are the gravity wave detectors that have been established around the world going to detect?

I'm still trying to find out what these gravity waves do.

MC Master of Ceremonies
08-12-2003, 08:54 PM
Originally posted by David Simmons
I'm going to take this as saying that my crude statement that once the pattern of "bent space" is established that pattern, or gravitational field, is stationary with respect to the mass that established it.

OK. So what are the gravity wave detectors that have been established around the world going to detect?

I'm still trying to find out what these gravity waves do.

Yes, your right, warped spacetime is basically the way that gravity acts, gravity waves are essientally waves through spacetime.

Gravity wave detectors are built to detect events like supernovae.

Whack-a-Mole
08-12-2003, 08:58 PM
Originally posted by David Simmons
When the earth passes between the sun and the moon the effect of the sun's gravity on the moon, and vice versa, is undisturbed. Gravity waves must not interact with matter. At least in that case.

Gravity certainly interacts with matter in that it attracts it but I think you are talking about blocking it.

As I remember it of the four fundamental forces (strong nuclear, weak nuclear, electromagnetic and gravity) gravity is the only one you can't shield yourself from.

The 'bent space' (like a bowling ball sitting on a rubber sheet) is usually the easiest way to look at it for us physics neophytes (although the ball on a sheet analogy isn't perfect).

David Simmons
08-12-2003, 10:57 PM
Originally posted by MC Master of Ceremonies

Gravity wave detectors are built to detect events like supernovae.

Aha! The vital clue! I think the light is beginning to dawn. In a supernova a large fraction of the mass of the star is converted into energy. So the gravitational field in the space right near the event changes abruptly. This change is then propagated at the speed of light sort of like the wave you can generate in a rope.

I was trying to form a mental picture of alternating gravity waves generating the static gravitational field and it just didn't work.

This also answers the question about how "gravity waves get out of a black hole." They don't and they don't have to as was explained by the link that was provided in an earlier post.

That's good enough. To form a deeper understanding I'd probably have to go back to school and learn all about tensor analysis and I have no intention of doing that. It's much too late.

scr4
08-13-2003, 12:06 AM
Actually a supernova may not emit much gravitational wave, as far as I know.

A gravitational wave is emitted when you move a mass. It's analogous to electromagnetic waves. When you grab a charged particle and shake it back and forth, this creates a disturbance in the electromagnetic field which propagates at the speed of light. This is how an antenna works - a high frequency AC voltage is applied to an antenna, and this causes the electrons to oscillate along the antenna.

So just shaking your fist creates a gravitational wave, albeit an infinitesimally weak one. A binary star system is constantly emitting a gravitational wave too. Ordinary star systems are large and have long orbital periods so the emitted power is small, but if you have a pair of neutron stars orbiting very close to each other the system loses a lot of of energy through gravitational wave emission. The effect was actually measured on the binary pulsar system PSR 1913+16 (http://astrosun.tn.cornell.edu/courses/astro201/psr1913.htm) and shown to match the value predicted by general relativity. Not only did this prove the existance of gravitational waves, it's also an important test of general relativity. The discoverers got a Nobel prize for it.

You'd think a violent explosion like a supernova would emit a strong gravitational wave, but that's not necessarily so. If the explosion is symmetric, there is no disturbance of the overall gravitational field. An asymmetric explosion would emit a wave, but it's not clear how asymmetric they are.

Maybe not a perfect proof but that's when governments started funding gravitational wave projects - before that people had doubts that there was anything to detect.

Achernar
08-13-2003, 12:10 AM
I heard that black hole mergers might produce enough gravity waves to be detected over cosmological distances. Is that true?

David Simmons
08-13-2003, 01:13 AM
Originally posted by scr4
Actually a supernova may not emit much gravitational wave, as far as I know.

Hmmm. Surely, if an electric charge suddenly changed from two electrons to one electron a detectable transient change in the magnitude of the electric field would occur. Since the electric field from the charges occupies all of space, that change in field strength would occur in all of space, but I don't think that it could happen everywhere at once so the change must propogate at the speed of light.

And isn't a large fraction of the mass of a star converted to energy in a supernova which changes the magnitude of the gravitational field, etc., etc., etc.

Gravitational waves seem fundamentally different to me than electromagnetic waves. For example, in a gravitational wave, what different field vector is propagated at right angles to the gravitational field vector?

This is even more complicated than I thought. And I never did think it is simple.

08-13-2003, 01:21 AM
(A pitiful earthbound creature seems to poke fun at those who seem to understand "gravity waves." Tittering ensues. Fingers point at AskNott. "Fen twitter tezz kile meftone," says he, clinging to his shredded dignity. "Telten yine foan." Oon yelleh mon. )

MC Master of Ceremonies
08-13-2003, 07:17 AM
Originally posted by scr4
Actually a supernova may not emit much gravitational wave, as far as I know.

A gravitational wave is emitted when you move a mass. It's analogous to electromagnetic waves. When you grab a charged particle and shake it back and forth, this creates a disturbance in the electromagnetic field which propagates at the speed of light. This is how an antenna works - a high frequency AC voltage is applied to an antenna, and this causes the electrons to oscillate along the antenna.

So just shaking your fist creates a gravitational wave, albeit an infinitesimally weak one. A binary star system is constantly emitting a gravitational wave too. Ordinary star systems are large and have long orbital periods so the emitted power is small, but if you have a pair of neutron stars orbiting very close to each other the system loses a lot of of energy through gravitational wave emission. The effect was actually measured on the binary pulsar system PSR 1913+16 (http://astrosun.tn.cornell.edu/courses/astro201/psr1913.htm) and shown to match the value predicted by general relativity. Not only did this prove the existance of gravitational waves, it's also an important test of general relativity. The discoverers got a Nobel prize for it.

You'd think a violent explosion like a supernova would emit a strong gravitational wave, but that's not necessarily so. If the explosion is symmetric, there is no disturbance of the overall gravitational field. An asymmetric explosion would emit a wave, but it's not clear how asymmetric they are.

Maybe not a perfect proof but that's when governments started funding gravitational wave projects - before that people had doubts that there was anything to detect.

No, supernovae do produce gravity waves and they are one of the dynamic massive systems that these detcetors are set up to detect.

MC Master of Ceremonies
08-13-2003, 07:18 AM
Originally posted by Achernar
I heard that black hole mergers might produce enough gravity waves to be detected over cosmological distances. Is that true?
yes coalesing black holes and binary neutron stars are other events that are expected to produce gravity waves.

scr4
08-13-2003, 08:01 AM
Originally posted by MC Master of Ceremonies
No, supernovae do produce gravity waves and they are one of the dynamic massive systems that these detcetors are set up to detect.
Did you even read what you quoted? I didn't say supernovae never produce gravitational waves. I said a perfectly symmetric explosion would not. We suspect that most real-life supernovae would, but that's hardly a proven fact. According to a review article on LIGO (http://www.ligo-wa.caltech.edu/P980007-00.pdf) (PDF file):
Supernovae have all the attributes associated with a good gravitational wave source: they weigh several solar masses, they are compact and they experience large accelerations. However gravitational radiation only couples to a changing quadrupole moment and, hence, if a supernova collapse and the subsequent explosion have an axial symmetry, no gravitational waves are emitted. There are several possible mechanisms which could overcome this deficit:
Initial density and temperature fluctuations may trigger the collapse unevenly, High rotation speeds can lead to a bar instability,
Hydrodynamic instabilities could introduce large convection streams which may effect the initial implosion
A reminiscent neutron star may experience a strong boiling shortly after its formation.
It is unlikely that each and every supernova event will be exactly axis symmetric, but how large the asymmetries are and how often these asymmetries lead to detectable gravitational waves is very much uncertain at this time.

As for merging black holes, IIRC with the next-generation instrument we can expect to observe them up to several hundred MPc (mega-parsecs). Not exactly cosmological distance. That's for stellar mass black holes - I don't know about supermassive black holes.

scr4
08-13-2003, 08:16 AM
Originally posted by David Simmons
And isn't a large fraction of the mass of a star converted to energy in a supernova which changes the magnitude of the gravitational field, etc., etc., etc.
Mass is not really "converted" into energy. Mass and energy are the same thing, and both create a gravitational field. A rock with mass M creates the same gravitational field as M*c2 worth of photons confined to the same volume or a flywheel spinning fast enough to contain that much energy.

In any case, the amount of mass converted into photons and other radiation is negligible compared to the mass of the star. A typical supernova explosion releases about 1051 erg of energy, which is 1030 grams. That's 0.05% of the mass of our sun.

I think you are correct that splitting of an electron into two would be observable. I should have said "axially symmetric."

David Simmons
08-13-2003, 08:53 AM
Originally posted by scr4
Mass is not really "converted" into energy. Mass and energy are the same thing, and both create a gravitational field. A rock with mass M creates the same gravitational field as M*c2 worth of photons confined to the same volume or a flywheel spinning fast enough to contain that much energy.

In any case, the amount of mass converted into photons and other radiation is negligible compared to the mass of the star. A typical supernova explosion releases about 1051 erg of energy, which is 1030 grams. That's 0.05% of the mass of our sun.

I think you are correct that splitting of an electron into two would be observable. I should have said "axially symmetric."

I think I was confusing the blowing off of a lot of the star's matter with it's appearance as energy. So if the explosion is reasonably symetric, the matter is still there and its center of mass is in the same place and the field should be relatively undisturbed.

However, there might be enough asymetry in some supernovas to generate detectable waves.

I guess I now begin to get a better picture of gravitational waves. They are generated by changes in the field and are not the thing that generates the field - correct?

And because all of the planets accelerate in their orbits the solar system is gradually radiating away its gravitational energy. Even if the sun by some weird happenstance didn't go nova, the solar system is still doomed. And the universe is flat and will just keep getting colder and colder through continued expansion and eventually all will be at the same temperature and nothing will happen.

I think I'll get into bed and pull the covers over my head.

Achernar
08-13-2003, 09:13 AM
Originally posted by David Simmons
And because all of the planets accelerate in their orbits the solar system is gradually radiating away its gravitational energy. Even if the sun by some weird happenstance didn't go nova, the solar system is still doomed.I wouldn't worry too much about this. The timescale for decay of the Earth-Sun system is on the order of:

5 c5 r4 / (64 G3 MSUN2 MEARTH) = 4 &times; 1027 years

Ring
08-13-2003, 10:19 AM
If you move a mass around the gravitational effects propagate outward at the speed of light and I suppose you could say these are gravitational waves.

However when physicists speak of gravitational waves they are normally referring to gravitational radiation, which is an entirely different animal. Electromagnetic radiation can be produced by a dipole moment because charge can be both negative and positive, and a dipole moment is very good at producing radiation.

However, since, as far as we know there is no negative mass, gravitational radiation can only be produced by a quadrupole or higher moment. And this involves the mass changing shape which is why scr4 says a supernovae explosion would have to be asymmetric in order to produce radiation. Also the quadrupole moment is not very good at producing radiation and this is why gravitational radiation is so hard to detect.

An intuitive way of showing how only a quadupole moment can produce G radiation is by invoking conservation of momentum. In order to make a mass (A) wobble between A1 and A2, there must be something causing the wobbling.... To dip back into Newton, the thing that is doing the wobbling must vary its momentum exactly opposite to what (A) does (action and reaction are equal and opposite) - hence canceling the dipole term via conservation of momentum.