How does an experiment at LIGO actually work?

How does an experiment at LIGO actually work?
The 4km long arms of the LIGO experiment at Hanford. Credit: LIGO lab:, Author provided

Gravitational waves are mysterious ripples in the fabric of space and time that travel across our universe at the speed of light. Predicted by Einstein exactly 100 years ago, a number of experiments have been searching for them. One of these experiments, LIGO, has been the focus of much speculation. But how does it actually work and how reliable is it?

Gravitational waves are caused by violent astrophysical events, involving massive, compact objects like neutron stars and black holes, colliding into each other. Even though the events that cause them are cataclysmic, they are so far away that the effects on our local fabric of space and time here on Earth are very subtle.

For that reason, scientists have had to build huge optical instruments that are supremely sensitive, called laser interferometers, to search for them. The Laser Interferometric Gravitational-Wave Observatory, or LIGO, brings these efforts together in an experiment with over 1,000 scientists from 86 institutions around the world working with these instruments or the data that they produce.

Two light beams, some mirrors and a detector

All you need to build a gravitational-wave interferometer is two , travelling between pairs of down pipes running in different directions, say north and west. The effect of a passing gravitational wave should stretch space in one direction and shrink it in the direction that is at right angles. On Earth, that would cause the mirrors to swing by tiny amounts, so that the distance between one pair of mirrors gets smaller, while the other gets larger. The swinging is actually the mirrors responding to the stretching and compression of space-time, which is just amazing.

How does an experiment at LIGO actually work?
A ring of particles influenced by a gravitational wave.

It is very similar to waves on a pond. Put down a floating object and, as a wave passes through, the object bobs up and down several times. LIGOs mirrors are bobbing in a pond of , which are more complex but still cause the motions to differ from place to place in a characteristic way.

The subtle changes in distance can then be registered by a , put in place to monitor the returning from the two interferometer arms. Just to ensure that it wasn't a fluke, we have two of these machines and position them at opposite ends of the US and require both of them to do the same "dancing mirrors" thing at the same time: one in Livingston, Louisiana and the other in Hanford, Washington.

So, how does this work in practice? A key task is "locking" the interferometers, which means stabilising the separations between the mirrors so that the laser light resonates between the mirror surfaces as it was designed to do. Back when I worked on a LIGO prototype at MIT in 1997, locking was done by hand by scientists, using a hand held box with 12 knobs on it. It is now computer-controlled, so that an operator initiates the sequence, and sensors indicate when each of the mirrors has moved into the right position. Mirror positions and angles tend to drift slowly due to temperature changes, mechanical relaxations in the hardware, and even the position of the moon in the sky, so adjusting the mirrors is a daily task.

How does an experiment at LIGO actually work?
LIGO control Credit: wikimedia

Scientists and engineers on-site also monitor diagnostic information about the detector and the physical environment, so that when the detector isn't working properly, the cause can be identified and addressed. I've spent many hours in the LIGO control rooms and labs; my most recent machine work was making precise measurements of the distances between mirrors during a troubleshooting exercise. In practice, this meant hours wearing clean-room garments and leaning over steel tables in a very large room, often working late into the night.

If I make this sound easy, it isn't. LIGO is oozing with groundbreaking technology developed especially for the detectors. The interferometer arms, each 4km long, had to be constructed with a correction for the curvature of the Earth. Each detector has to be exquisitely isolated from vibrations of the ground and it has to be in a vacuum so that contaminants and gas don't corrupt the laser light between the mirrors.

The two detectors have to take data for months at a time – never missing a single data point and never getting behind. When your detector is distributed over several kilometres, this is a technological challenge in itself. LIGO is an engineering and physics marvel, one of the most sophisticated machines ever constructed and it's exciting to be part of it.

Explore further

Announcement Thursday on Einstein's gravitational waves

This article was originally published on The Conversation. Read the original article.
The Conversation

Citation: How does an experiment at LIGO actually work? (2016, February 11) retrieved 18 September 2019 from
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

Feedback to editors

User comments

Feb 11, 2016
SO Ed, how is the interferometer isolated from Spacetime so that the measuring device doesn't experience the oscillation it is supposed to measure?

Huh? If it were isolated from spacetime (whatever that means) it COULDN'T measure what it is supposed to measure. The thing measures effects due to warping of spacetime.

OK...but you're instrument is trying to measure the vertical "bob" by constantly checking the distance between the instrument and the waters surface, since the object moves with the waters surface that distance NEVER changes

Even for the water analogy (which is a bad analogy) that assertion isn't correct. You can try it out in your bathtub. When one end is already within the effect of the wave while the other end is not then you get a difference in length.

Feb 11, 2016
In order to measure wind velocity, the instrument doing the measuring can't be carried by the wind it is trying to measure,

Good thing then that garvitational waves don't move with the instrument but through it, eh?

Feb 11, 2016
Go to the press conference (currently running...but will surely be on youtube in its entirety afterwards). They explain it very well how they did it (I guess I'm not spoiling it at this point that gravitational waves have been detected)


Feb 11, 2016
I wrote up a short intro to the subject, it's here:

Feb 11, 2016
It's funny that you have no idea what that means. What precisely makes you think anything outside the reference frame of the wave will move because of the wave? The theory is a ripple in spacetime itself right? Not a ripple in the matter.

Go to the video of the press conference if you don't understand what I'm saying. They explain it at length. Even some pretty pictures for you to look at.

You can cry "no" all you want...but experiment trumps your belief...every time.

Feb 11, 2016
How does an experiment at LIGO actually work?

By detecting the drag of the ether with lasers and mirrors.

Oh wait that was Michelson and Morely's test case for it. Oh well, glad to see it works in a much more refined form :)

Feb 11, 2016
I don't care wtf anyone says. 1/1000 of a proton can't be measured.
The measuring device is made of protons. The wavelength to measure a 1/1000 of a proton is minimum 1/500 of a proton which as never been measured either.

We are talking attometer wave length people.

Feb 11, 2016
Nonsense, it's a space-time wave, even if it existed you would not notice any change in distance due to the like change in time. But why argue the unarguable. This is a belief system. The wave front is not a constant. i.e, no optical connector can deliver a multi nanosecond delay with a clear signal. The distance into the fiber is not that far. Over a meter? Get real! Lies!!! Not physics.

Feb 12, 2016

We are talking attometer wave length people.

no optical connector can deliver a multi nanosecond delay with a clear signal.

Just google for attometer spectroscopy. You'll find no shortaged of papers/experiments that demonstrate how to do this.

Feb 12, 2016
Is this experiment fundamentally different from it's first implementation?

Feb 12, 2016
You can believe what you want to believe.

I tend to favor evidence over belief. That's just scientific. If you have evidence to the contrary, present it. But the arguments you have brought forth in this and other threads on the gravitational waves are just plain silly.

you are supporting someones interpretation of an event that has no other means of verification

The experiment was done to test the hypothesis. That's what you do in science. And yes: The experiment will be repeated by them and others to check if it was a fluke, never fear. They are not going to shut down LIGO tomorrow.

An event theoretically caused by a merger between 2 theoretical objects that produced no verifiable signature in the EM spectrum

Use your head: From what mechanism WOULD you expect a signature in the EM spectrum during such a merger?
Black holes don't radiate unless they are in the feeding stage - and even then you only see an EM burst in certain directions.

Feb 14, 2016
I question the accuracy of our knowledge of the distance (1.3 million ly) with enough accuracy to correlate / attribute the source to a resolution of the source of the wave within the three months that the instrument has been working in it's upgraded sensitivity. Possibly it could have been shift in mass magnitude or position at a much closer distance to generate the same amplitude wave. We need to at least know the statistics of such measurement before attributing the source of the event.

Feb 14, 2016
I question the accuracy of our knowledge of the distance (1.3 million ly)

In the press conference they said that there was some uncertainty in the distance (1.3 billion ly is just the most probable. It could be 900 million ly or 1.6 billion ly)

But from the matching of the signal with the model you get the masses - and from that you get the amplitude of the signal at the source. From that you get the distance based on the measured amplitude.

It's like with standard candles: if you know how bright they appear you know how far away they are. Since there is not much variation in black holes -as they are only characterized by mass, spin and charge- you have very little variability that can throw a measurement.

If all else fails read the paper.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more