Can LIGO actually detect gravitational waves?

[swansont]

"Change" implies detectable change. So if the arms are waving to exactly the same degree as the space they occupy (as they must under the definitions of these terms in this context), then it's entirely undetectable.

It's only undetectable if you use a ruler of some sort. So we don't use a ruler.

 

So the lengths of the arms are not changing.

No.

 

Which brings us back to my initial question: what is being measured?

The length changes, as measured by something that depends on time rather than length. If the length changes but the speed of the probe is constant, what happens to the time?

 

So how in this case is the speed of light being changed by grav waves, and in such a way that any change in speed would be detectable by an interferometer?

The speed of light is not changing. This has been mentioned by several people in the discussion. I don't know why you think it is; the local value is always c.

[aramis720]

It's only undetectable if you use a ruler of some sort. So we don't use a ruler.

 

 

No.

 

 

The length changes, as measured by something that depends on time rather than length. If the length changes but the speed of the probe is constant, what happens to the time?

 

 

The speed of light is not changing. This has been mentioned by several people in the discussion. I don't know why you think it is; the local value is always c.

We're going in circles here. The speed of light is invariant with respect to what? I understand, of course, that c is invariant, but various parties have offered here that it is the invariant speed of light that gives rise to detectable effects from the interferometer. I'm asking how this is the case. If you assert that the interferometer can detect grav waves b/c of a difference in phases of the two light signals sent up and back in each arm, there must be a change in wavelength or length of the arms or both. Most here have agreed that there is no change in either, so what will happen? No difference in phase overlaps of the two light signals. "Time" in this context is inferred from change in phase fringes, but there can't be any change in phase fringes based on the definitions offered in this context.

Edited June 5, 2017 by aramis720

[Strange]

I could have stated more clearly my summary of your previous statements, I'll agree on that, but my point remains. You had agreed that length and wavelength change to the same degree exactly as the wave, and thus are undetectable, that was my point. You now clarify that because the speed of light remains constant that the interferometer will register a difference. But think again about what is going on with the measurement apparatus. Light does not exist in some other realm than the space we exist in. So if space itself is being distorted so is any light occupying that space. You write:"Because if light takes a different amount of time to travel one arm than the other, it will have a different phase when it gets back." But why would it take a different amount of time to travel one arm? We know, of course, that light is affected by gravity in the same way as mass -- this was the basis for the famous 1919 Eddington experiment looking at the curving of light around the sun during a full eclipse.

 

 

And the fact that light is affected (more in one arm than the other) is exactly why it works.

[tar]

aramis720,

 

While I have the same problem as you do, trying to figure out what stays the same and what varies here, one way to look at it, is to think backward, as in what would cause a beam to leave at one frequency, get split, go down one arm and come back blueshifted at the same time as the other half of the beam comes back from the normal arm redshifted or vice-a-versa?

 

Regards, TAR

Edited June 5, 2017 by tar

[swansont]

We're going in circles here. The speed of light is invariant with respect to what? I understand, of course, that c is invariant, but various parties have offered here that it is the invariant speed of light that gives rise to detectable effects from the interferometer. I'm asking how this is the case. If you assert that the interferometer can detect grav waves b/c of a difference in phases of the two light signals sent up and back in each arm, there must be a change in wavelength or length of the arms or both. Most here have agreed that there is no change in either, so what will happen? No difference in phase overlaps of the two light signals. "Time" in this context is inferred from change in phase fringes, but there can't be any change in phase fringes based on the definitions offered in this context.

The difficulty I had in figuring this out was I was thinking of the light referenced to itself (or the test mass), and that isn't the comparison we are doing. So it doesn't matter that both the arm and the wavelength are adjusted. All that matters is the phase of the returning light, and it is changing with respect to the other arm.

 

I think tar's comment is valid: the light in one arm is blue-shifted and the other red-shifted with respect to "normal". How can you not have a change in the interference?

[Mordred]

The difficulty I had in figuring this out was I was thinking of the light referenced to itself (or the test mass), and that isn't the comparison we are doing. So it doesn't matter that both the arm and the wavelength are adjusted. All that matters is the phase of the returning light, and it is changing with respect to the other arm.

 

I think tar's comment is valid: the light in one arm is blue-shifted and the other red-shifted with respect to "normal". How can you not have a change in the interference?

Tar's response is valid and quite accurate.

 

Section 5 of the this LIGO paper has some of the applicable formulas.

 

see section 5

https://www.google.ca/url?sa=t&source=web&rct=j&url=https://physics.aps.org/featured-article-pdf/10.1103/PhysRevLett.118.183602&ved=0ahUKEwiX5b6G0afUAhVmrFQKHafWBDMQFggdMAA&usg=AFQjCNHj5HzdaiCXdKwtT_I3zLkDR4_z-Q&sig2=-SnqX9aZTHKHkTLycIW71A

 

It gives the phase variation formulas due to GW waves. Took me a bit to find a reference specific to the topic with the applicable formulas.

 

Actually here is the generic formula involved. This article is more detailed on the actual metrics.

 

REFERENCE FRAMES AND THE OBSERVABLE PHASE SHIFT OF THE CARRIER IN G.W. INTERFEROMETRIC DETECTORS

 

https://www.google.ca/url?sa=t&source=web&rct=j&url=http://www.pas.rochester.edu/~meliss/URLIGO/030428_Referenceframes.pdf&ved=0ahUKEwiX5b6G0afUAhVmrFQKHafWBDMQFggqMAM&usg=AFQjCNEP36c0mLAulF75nU3g5SO1_hzfow&sig2=w-TEWmvM-A9cgmz1TfG12g

 

equation 15.

Edited June 5, 2017 by Mordred

[Strange]

I think tar's comment is valid: the light in one arm is blue-shifted and the other red-shifted with respect to "normal". How can you not have a change in the interference?

 

 

That is a nice way of looking at it. And it clearly relates to aramis720's comment that "We know, of course, that light is affected by gravity..." And... one of those effects is red/blue shift.

[Mordred]

Wow no matter how many times I read the Calibration papers on LIGO it never ceases to amaze me how incredibly accurate and sensitive it is.

 

Using 10 lead bricks, placed at 25 cm to the interferometer to calibrate for 1 rad.

 

😲 some details from the first article. Rather interesting what they have to filter out to establish a baseline.

Edited June 6, 2017 by Mordred

[tar]

Mordred,

 

Makes me wonder if some if not most of the "noise" we throw out, is actually signal.

 

I am thinking again of the chirp we "heard". We are not listening for long slow rumbles or hums of various pitches as we probably should be. The periods we should be listening for are not to be measured in milliseconds, but in days and months and years and centuries and millennia and eons.

 

A different attitude toward the data might be in order. Don't "look" for a pattern. Just look at ALL the data, and "see" the patterns that actually exist.

 

It might be analogous to a cloudy day. Even though you are not going to see the sun, or the moon, it is still going to be light outside and the tides are still going to cycle.

 

Regards, TAR

[Mordred]

Its possible but recall an earlier point I made on types of polarity. Mechanical vibration and electro magnetic signals have different polarity influence to a GW wave. So from research of the reviewed signals they have received they can develop a statistical means of filtration.

 

As more confirmed GW data develops they will also be able to better hone in on key aspects of a GW signal to better identify them.

 

When you have a sensor looking for signals that in essence rely on 10^-20 of a cm in length change you can certainly see the difficulties of filtration.

[imatfaal]

Its possible but recall an earlier point I made on types of polarity. Mechanical vibration and electro magnetic signals have different polarity influence to a GW wave. So from research of the reviewed signals they have received they can develop a statistical means of filtration.

 

As more confirmed GW data develops they will also be able to better hone in on key aspects of a GW signal to better identify them.

 

When you have a sensor looking for signals that in essence rely on 10^-20 of a cm in length change you can certainly see the difficulties of filtration.

 

Which is why you have two such sensors situation thousands of miles apart and compare results. Both will see lots of variations of that magnitude in noise and the noise will be unique to the environment of the sensor - but only the signal will be reproduced in both

[tar]

Mordred,

 

Understood, but not looking to filter anything out.

 

Imagine for instance how many "things" you have noticed by studying Google Earth. Nobody is filtering out trees because you are looking for railroad tracks.

 

You can't see the forest without being able to see the trees.

 

And there is no way, from the pictures to discern whether a feature is natural or manmade. You have to combine your senses, and your memory and the information from various sources to tell what a thing is. Like the lines on the bottom of the ocean created from sounding data gleaned from a particular ship on a particular course, the actual shape of the bottom has to be imagined by filling in the blanks between the lines.

 

Imagine just taking the data and finding some way to display it in 3D 360 degrees, and overlay the display over an actual 180 degree view above the detector and superimpose a view taken from the opposite point on the Earth. Then you could "see" what to associate with what.

 

Regards, TAR

 

I am thinking of using the twelve sections of the sphere idea to create a planetarium idea both above and below. Up would be of course up from wherever the planetarium is located, but every panel on the inside surface of the sphere would be exactly associated with that direction from the center of the Earth. Any real time information from camera or telescope or radio telescope could be displayed. Once three interferometers are online, the triangulated direction of any and all signals can be displayed.

[swansont]

Mordred,

 

Makes me wonder if some if not most of the "noise" we throw out, is actually signal.

 

I am thinking again of the chirp we "heard". We are not listening for long slow rumbles or hums of various pitches as we probably should be. The periods we should be listening for are not to be measured in milliseconds, but in days and months and years and centuries and millennia and eons.

 

 

 

My understanding is that the frequency and the amplitude are correlated. The longer-period signals will be correspondingly smaller in amplitude. Since it would be beyond the sensitivity capability of the instrument (of existing or imagined experiments), there is no point in looking.

[tar]

Well, then if the only thing we can hear is black holes merging a billion years ago, what's the point?

if we can hear a chirp as a tiny thing a billion lys away spins, why can we not theoretically hear a big lumpy thing humming close by?

amplitude, I would think is also correlated to distance from the source

Edited June 6, 2017 by tar

[DrP]

Oh, yea, 'Only' 2 black holes merging a billion years ago... what's the point eh? lol

 

Personally I think it's incredible, amazing... words don't describe the wonder even. I am sure we will learn a lot from this.

[tar]

DrP,

 

My statement was not meant to say that detecting a gravitation wave was not important. My thought was merely that if a tiny thing a billion lyrs away shouting can register, then that implies a big thing close by would also bend space, but at a different frequency and amplitude, and if only a tiny really dense pair of black holes can even generate a wave, what does detecting that wave do for us. If we are to learn anything by it, I would imagine that how to sense gravitational ripples would be one of the things. If we can never hope to sense anything but the loudest shouts from small dense objects from way long ago, and way far away, then what utility does it have for our survival?

 

Regards, TAR

...really tiny things spinning around each other, bending space a billion light years away...an electron dropping a quantum piece of energy in the person of a photon that bends the electric and magnetic fields for a billion lys to get detected by the sensor at the end of our telescope.

 

I once heard that the light of a match on the surface of a new moon, would put off enough light to be detected by a human eye on Earth.

 

Why would it be impossible to pick up a gravity wave generated by an orbiting moon of Jupiter?

Edited June 6, 2017 by tar

[imatfaal]

Tar

 

"My thought was merely that if a tiny thing a billion lyrs away shouting can register, then that implies a big thing close by would also bend space, but at a different frequency and amplitude, and if only a tiny really dense pair of black holes can even generate a wave, what does detecting that wave do for us."

 

the above is wrong - it was not tiny. In the final stages of the merger the two black holes gave off energy at a higher rate in gravitational waves than all the light from all the stars in the observable universe

[tar]

That size thing brings up a point that I am wondering about. The computer simulations of the two black holes that merged that produced GW150914 showed two giant masses, black holes with stars and galaxies around them. As in really big. If the last moments before merger each of the pair was moving at .6C that means they must have been really small to get around each other. Not the size of two galaxies that would take a quarter million years to circumvent each other at the speed of light, but the size of small planets, or basketballs or something. We don't have any instruments that could resolve such small masses into two masses to begin with. Why would we consider the thing to be optically two things?

very energetic, and very big are not the same thing

imatfaal,

 

But if my eye can detect a couple photons out of the trillions a single star puts off in my direction, and that star puts off trillions in all directions, and all the energy output of all the stars, is just a fraction of the energy put out in one wave at black hole merger in the form of a gw, then the merger should have indeed shook the place.

 

But how much energy is put off, in the form of gravitation wave, when a huge solar flare jets out of a rotating Sun.

 

Enough I would think, for a properly constructed gw detector to pick up. If my eye can pick up an electron dropping an energy level in a hydrogen atom 4 lys away, that is.

 

Regards, TAR

[swansont]

Well, then if the only thing we can hear is black holes merging a billion years ago, what's the point?

if we can hear a chirp as a tiny thing a billion lys away spins, why can we not theoretically hear a big lumpy thing humming close by?

amplitude, I would think is also correlated to distance from the source

 

 

 

A black hole is not tiny. Merging black holes are one of the more violent interactions; the first detected even has a conversion of something like 3 solar masses into gravitational energy during the detectable event. Again, not small.

 

Longer-period interactions would be orbits, and they shed a lot less energy than a merger. I remember a talk a couple of years ago by Joe Taylor (of the Hulse-Taylor binary pulsar Nobel prize) He explained how lucky they were to have found/studied that particular system, because only a few binary pulsars have orbits that would lend themselves to detecting the orbital changes that provide evidence of emitting gravitational radiation, and that took place over much longer times than tenths of seconds. (amplitude at earth would be about a part in 10^26, or 5 orders of magnitude smaller than what was detected in the first LIGO event, and is only ~21k LY away as compared to a billion) Less emitted energy per orbit and longer periods mean there are two factors conspiring to lower the power of the signal you wish to detect. Both of those factors are smaller by many orders of magnitude.

 

The local effects are a part in 10^25, or 4 orders of magnitude smaller than what was detected in the first LIGO event, and, obviously, a heck of a lot closer

https://en.wikipedia.org/wiki/Gravitational_wave#Wave_amplitudes_from_the_Earth.E2.80.93Sun_system

 

The only thing in your favor here is that the radiation drops off with r instead of r^2, since it's a quadrupole

http://www.tapir.caltech.edu/~teviet/Waves/gwave.html

 

You might also note in that link that the gravity gradients that give rise to the radiation vary with f^4, so when frequency drops by a factor of 10, the gravity gradient drops by a factor of 10,000

That size thing brings up a point that I am wondering about. The computer simulations of the two black holes that merged that produced GW150914 showed two giant masses, black holes with stars and galaxies around them. As in really big. If the last moments before merger each of the pair was moving at .6C that means they must have been really small to get around each other. Not the size of two galaxies that would take a quarter million years to circumvent each other at the speed of light, but the size of small planets, or basketballs or something. We don't have any instruments that could resolve such small masses into two masses to begin with. Why would we consider the thing to be optically two things?

 

That's an issue of resolving power of optical instruments and the distance. You might not be able to resolve two distant trees as one with a set of binoculars, but that does not mean there aren't two trees.

 

The Schwarzschild radius of a BH of 30 solar masses is ~ 90 km. Your idea of scale is off, I think.

 

http://hyperphysics.phy-astr.gsu.edu/hbase/Astro/blkhol.html

 

But how much energy is put off, in the form of gravitation wave, when a huge solar flare jets out of a rotating Sun.

 

 

It can't be more than a tiny, tiny fraction of the mass of the sun, because the sun is still there afterwards. And in terms of gravitational radiation, you have to look at the quadrupole component — a gravity gradient must exist for the radiation to be given off

[tar]

"In the latest LIGO event, a black hole 19 times the mass of the sun and another black hole 31 times the sun’s mass, married to make a single hole of 49 solar masses. During the last frantic moments of the merger, they were shedding more energy in the form of gravitational waves than all the stars in the observable universe".

 

 

Is a quote from an article on the June 2017LIGO event. I am thinking that this in conjunction with the small wattage usually put out by rotating masses in the form of gravitational waves points to a possibility that imatfaal's quote about equal to all the energy of all the light from all the stars in the observable universe might be a misunderstanding. It appears that the energy put off by a star in the form of gravitational waves is very much smaller than the energy put off by the same star in the form of electromagnetic radiation.

still, the size thing gets me if two things get around each other in a second and the speed of light is 186,000 miles a second, and the thing at the last instant before merger is going .6C that means in a second it went at most 111,600 miles in a circle. the Dec event the holes orbited 27 time in the last second before merger, meaning the circumference of the orbit was 4133 miles

 

Divided by pi for a diameter that is 1,316 miles as the max diameter of either hole. Despite the mass being huge, equal to scores of suns, the size was tiny. An if you add in the size of the event horizons of each of the holes, there could be no optical differenciation between the two, prior merger, so why do the computer simulations show optically, two holes? And if the two masses were so close to each other anyway, what is the difference between prior merger and a spinning mass after merger, in terms of gravity? That is, what is so special about the merger that the act should shed 3 solar masses worth of energy in the form of gravity wave?

Edited June 6, 2017 by tar

[swansont]

still, the size thing gets me if two things get around each other in a second and the speed of light is 186,000 miles a second, and the thing at the last instant before merger is going .6C that means in a second it went at most 111,600 miles in a circle. the Dec event the holes orbited 27 time in the last second before merger, meaning the circumference of the orbit was 4133 miles

 

Divided by pi for a diameter that is 1,316 miles as the max diameter of either hole. Despite the mass being huge, equal to scores of suns, the size was tiny.

 

As I already noted above.

 

An if you add in the size of the event horizons of each of the holes, there could be no optical differenciation between the two, prior merger, so why do the computer simulations show optically, two holes?

No optical differentiation at some distance, as defined by the Rayleigh criterion. But the simulations do not assume you are a billion LY away, which should be obvious by the scale of the images.

[tar]

well, please explain this gravity gradient thing to me again, does it have to do with the density of the one hole as compared with the other, or the density of the solar flare as opposed to the Sun proper?

 

 

The distance from the holes, judging by the scale of the simulations of GW150914 would be just a few hundred miles...way to close to include whole stars and galaxies in the picture, so somebody's scale is off, but whether the two circling holes are a hundred miles apart, or 1000 miles apart, during their last second death spiral, the idea is the same, that their gravity is basically coming from the same tiny area of space, before, during and after merger. So where does the gravity gradient come from during the last milliseconds, that was not there a second earlier and a second later?

what happens at merger that turns 3 solar masses into gravity wave?

 

https://www.bing.com/images/search?view=detailV2&ccid=YNCiW0yZ&id=3A21F794B4983B244C5B9EFA984CB23B466AC320&thid=OIP.YNCiW0yZvHI2waY0PALFbwEsCo&q=GW150914&simid=608003693727713809&selectedIndex=4&ajaxhist=0

take the average distance between two stars as lets say a ly and look at the closest 2 and see how many times that distance would fit across the diameter of one of the simulated holes

Edited June 6, 2017 by tar

[swansont]

well, please explain this gravity gradient thing to me again, does it have to do with the density of the one hole as compared with the other, or the density of the solar flare as opposed to the Sun proper?

 

It's the tidal field (g'), i.e. the difference in gravity on two test masses at some separation. That has to be varying in time; the strain (what is measured in LIGO) is a double integral of g' with time

 

http://www.tapir.caltech.edu/~teviet/Waves/gwave.html

 

No tidal field (i.e. uniform gravity) will not give you radiation. A static tidal field will not give you radiation. A tidal field varying linearly in time will not give you the radiation, so simple linear motion toward a mass will not give you the radiation. You need an acceleration.

 

The distance from the holes, judging by the scale of the simulations of GW150914 would be just a few hundred miles...way to close to include whole stars and galaxies in the picture, so somebody's scale is off, but whether the two circling holes are a hundred miles apart, or 1000 miles apart, during their last second death spiral, the idea is the same, that their gravity is basically coming from the same tiny area of space, before, during and after merger. So where does the gravity gradient come from during the last milliseconds, that was not there a second earlier and a second later?

 

The simulations are showing you the background as you would see if you were in space some relatively short distance away. The galaxies and stars do not need to be close to the black holes; they give you no hint of the scale.

 

 

https://www.bing.com/images/search?view=detailV2&ccid=YNCiW0yZ&id=3A21F794B4983B244C5B9EFA984CB23B466AC320&thid=OIP.YNCiW0yZvHI2waY0PALFbwEsCo&q=GW150914&simid=608003693727713809&selectedIndex=4&ajaxhist=0

take the average distance between two stars as lets say a ly and look at the closest 2 and see how many times that distance would fit across the diameter of one of the simulated holes

 

Why a LY? And why would you assume that all stars in a picture are the same distance away from you? Is that ever true, for a picture into outer space?

[tar]

swansont,

 

The simulation is way misleading. There are big stars in the foreground and tiny stars further back giving an impression that you are looking at massive in diameter holes with clouds of stars around their perimeter. I would say it is meant to give the impression of a view from 10s of thousands of light ​​years...but hey its a simulation, not worth arguing over, it does not mean anything, there is no real components to it.

f

But back to the gravity gradient. The link gave me a 404 error so I am just going by your two test mass description, and I am not sure whether the test masses are far from the source of the gradient or the two masses creating the gradient. The strain on the space within the interferometer is what I thought we were measuring, and the distance between the mirrors is not exactly the distance of the path of the beam, because the beam is sent back and forth quite a few times before recombined. So the strain on the mirror at the end of the path and at the leveraging mirrors near the half silvered one, is different at any t only by the gradient between, not the full beam path distance implied gradient. All this is probably figured, but I am concerned as to what we figure the distance between the mirrors means in terms of what happened with the black holes, 1.3 billion years ago. That is, if the gradient between the mirrors squishes the space between the width of a partial proton, and that is because some huge mass was accelerating very close to another huge mass 1.3 billion years ago, wouldn't a comet making its close approach to the Sun cause a little chirp itself?

 

Regards, TAR

relativity wise does it matter if the Sun is M1 and the comet is M2? that is can you consider the Sun accelerating toward the comet?

Edited June 6, 2017 by tar

[tar]

well forget the comet question. Comets are just a little dust and gas, not massive at all. But when a planet orbits, I understand that it is in essence, falling around the Sun, or a moon falling around its mother planet. They are accelerating and therefore would have a non zero integral.