LIGO and 'Michelson-Morley' experiments
9 Oct 2006 21:19:30 UTC
Topic 191927
(moderation:
Just wondering - to what extent are GW detectors like LIGO similar to the appartus used in the various Michelson-Morley experiments?
Of course, there are many differences in things like the path lengths, the fringe-detection, the light sources, ... but in terms of the underlying physics, what are the key differences?
And if the key infrastructure were kept (tunnels, lasers, mirrors, etc), but the measurement/detection bits perhaps swapped out, how close to an MMX-type device could the LIGOs be?
LIGO and 'Michelson-Morley' experiments
)
Very close indeed! They are termed Fabry-Perot-Michelson interferometers.
What is extra is the technology advantages over 1887, and that each arm contains a form of Fabry-Perot cavity. This is basically two mirrors ( kilometers apart here ) held at a constant distance to within 1/2 wavelength tolerance. The laser light bounces back and forth many times in 'resonance' when this is achieved. There is an incredibly detailed control system to actuate the positions/movements of the components to obtain this. By reading the 'dark' port - which is the one where the 'crest' of the light phase from one arm cancels the 'trough' of the light phase from the other - then movements are made to keep that at maximum darkness. This is a more sensitive approach than leaving the separations undisturbed and reading the light variation. The signals within the system used to achieve this state are then used as a 'proxy' measure of any gravity wave passing. One then needs to calibrate, that is determine what control signal level corresponds to what level of spacetime distortion.
That output ( labelled AS_Q ) becomes the fodder for the data analysis pipeline of the E@H project.
Cheers, Mike.
( edit ) The interior of a laser is very similiar to a Fabry-Perot etalon. In a sense a whole arm is an extension of the laser resonant cavity which recycles the photons up and down the arm .......
but with no stimulated emission in the vacuum of the arms.
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal
Thanks Mike. How much more
)
Thanks Mike.
How much more sensitive is LIGO than the original MMX?
I know, for example, that the arm lengths are far greater, but what about the ability to detect fringe shifts (IIRC, MMX could observe ~1/4 of a fringe shift)?
And does the normal operation of LIGO mean that any 'aether' which MMX ruled out LIGO also rules out, to far greater accuracy?
RE: How much more sensitive
)
Quite a few orders of magnitude I'd think.
Several points here:
- LIGO's have access to far purer light than Michelson & Morley ( M&M ) did. I think they used an arc/vapour lamp ( I'm not sure ) - certainly with a far great spread of frequency than a modern CO2 laser, and more variation in intensity. LIGO's have state of the art lasers with power stability and 'mode cleaners' - optical filters/cavities which discourage transmission of unwanted photons.
- LIGO's have better detectors. In M&M's day it was a magnifying scope focussed on an image plane where the two beams recombined - thus read by the human eye. They rotated the detector as whole and compared any change in fringes between the two orientations ( hypothetically swapping each arm from parallel to perpendicular to the aether ). LIGO's have all the magic of modern solid state gear.
- LIGO's are better isolated from disturbance. While M&M floated their experiment on a mercury bath, the LIGO's have an incredibly well engineered set of apparatus to reduce that. This is called HEPI ( Hydraulical External Pre-Isolator ):
plus mounting to passively/actively damp the optical elements ( see here ). There is also a thermal compensation system to counteract the effects of mirror distortion - it does so by shining another laser on a mirror to alter it's shape by differential heating across the mirror surface:
- it's all done in high vacuum, with high cleanliness and purity. Equipment changes are quite exquisitely done ( see here ).
- as you say the light path distance over which a shift will become apparent is far greater. M&M did bounce the light back and forth across the table about half a dozen times before re-combining to increase that. The LIGO's, by using the Fabry-Perot cavities, in effect do a similiar thing when locked in resonance.
Hmmmm..... I suppose there could be two answers:
- no one is looking for that, as no-one needs to believe in an aether - Einstein considered it 'superfluous '. This means we can continue to ignore it as theory and practice, using Occam's razor, should be rid of it.
- with devil's advocacy let's assume the aether is there in a measurable sense, so what would we see? Taking the aether theory implying some preferred reference frame for measurement ( the 'at rest' one ) then some phenomena could declare the motion of another frame with respect to that unique frame. This presumably would happen with some velocity dependence ( direction and speed ) of our test frame. So as the Earth rotates around the North/South axis, with LIGO's on it, and moving in it's orbit around the Sun, and the Solar system rolling around the Galactic centre etc. We ought to hit an alignment of our test frame ( the LIGO ) with the universal aether frame ( the 'at rest' one ), or if not deduce our velocity with respect to it. If we then use the phenomenon of the propagation of light as our test case then we expect some variation of that as the frame orientation evolves due to the various motions described ( other disturbances eliminated or accounted for ) - a 'fringe shift' as you say. No-one has yet mentioned seeing an effect in that regard ... :-), so I guess it has been ruled it out, and to greater accuracy than M&M!
Cheers, Mike.
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal