0:27

Okay, so we restart DS9, go to Analysis>Virtual

observatory. Look at the primary MOOK.

Here's our set of possible observations. And let's

scroll down to obs ID

1943, the wind and accretion disk in Cen X-3.

And now you can see what Cen X-3 looks like in Chandra.

It looks very, very peculiar, almost like a solar eclipse here.

That's because Cen X-3 is so bright, and we're using kind of a different

detector for this observation than for a lot of the other Chandra observations.

But our green regions are already set down here in order

to capture most of the photons that are present

or are coming from Cen X-3 into our

satellite. So let's now look at our light

curve. Let's first get our analysis programs

set up in the way we usually do.

1:52

And we will look at our light curve just like we

did for the EXOSAT data. When

we click on Light Curve, our computer chugs

along and after a little bit of

time, here we go. Here's our

light curve. Also wildly varying up and down,

very similar to what it looked like in the EXOSAT observation.

Notice though, this is not as long an observation.

Okay, this is about 50,000 seconds.

3:01

Now we're getting a little bit closer to seeing what's going on.

We'll zoom in again.

We left-click and

then release. And, oh, look at that!

Boom, boom, boom, boom, boom.

You can count the peaks here and count the amount of time.

And sure enough, it looks like it's about 4.8 seconds again.

But is it exactly? Let's go to our Power Spectrum and

find out. We click on the Power Spectrum

3:42

and here it is: 0.2 seconds per, se-

se, 0.2 cycles per second. Let's zoom

in on it, left click, make our box here.

Here we go.

Do it again.

Left click and drag our box and there we have it.

Look at this! Our frequency for

our EXOSAT data was at about 0.207.

There's absolutely nothing in the Chandra observation at point 207.

It's moved! It's moved to 0.208 cycles per second.

4:34

Look! The frequency is slightly different.

We get a doppler shift range just like we did before,but

now with Chandra, it is centered on 0.2080

seconds, instead of, 0.2071 seconds. So

f for Chandra, it is centered

on a frequency equal 0.2080

cycles per second, or a period

of about 4.81 seconds

instead of our frequency

with EXOSAT, which was about

0.2071 cycles per

second, for a period with EXOSAT

of about 4.83 seconds.

There's a clear difference here. It's a small

amount, but look at the power spectra of

these two satellite observations side by side.

6:48

What this means is Cen X-3 has spun up in 15 years; it's going faster.

And in fact, when you look at it over even longer time spans, it seems relentlessly

and predictably gaining speed in a more or less linear fashion.

What could be causing that? Well, it appears

that the companion star feeds the x-ray source some material, and as that

gas gets closer and closer to the neutron star, it gives the star a bit of a kick.

It's quite similar to what happens when an ice skater

does a spin and draws in his or her arms.

They spin faster and faster and faster

to conserve angular momentum. Pretty neat.

7:45

One more interesting thing we can do is to

find out the luminosity of this object in the x-rays.

From the optical brightness and spectrum of Krzeminski's star, we have deduced that

the distance to Cen X-3 is about 20,000 light years.

Now, from the x-ray observations, we can find the average flux,

or the amount of energy that passes through each square

centimeter of our satellite's detector each second.

To do this, we return one last time to DS9

and examine Cen X-3's energy spectrum.

8:32

We go to Analysis, and now we're going to do our Chau

Sherpa spectral fit. This is going to take all the photons

in the observation and fit it to a particular model of radiation.

There are lots of different models that you can try

to fit your data to, and in this situation, it's not

really all that important, because we just want to see the overall number,

an amount of radiation that's passing through our satellite

detectors independent of the actual way that it's actually radiating.

So we're going to choose the McAll data fit, or the McAll spectral fit.

But what we do want is to display the log, because we are going to be interested

in the flux, the amount of energy passing

through each square centimeter of our detector each second.

So we click on Display Sherpa logs.

And now we wait for our computer to do this analysis.

It's fitting hundreds and hundreds of thousands of photons

and it's going to take a little bit of time.

But here we are, and now you can see this is the plot of the energy output of

Cen X-3 superimposed with a little white line that's almost impossible to see.

We don't have to worry about that.

That's the actual model fit. What we're interested in is this

number near the top of our log. If our model choice

is more or less valid, we can use this flux to

predict the intrinsic luminosity of the object.

This is the flux that we would get from Cen X-3

if there was no absorbing gas and dust

between Cen X-3 and the Earth. This

is the number we want, it's about 2.4 times ten

to the minus nine ergs per centimeter squared per

second. The flux is 2.4 times 10 to

the minus 9 ergs per square centimetre per second.

This means that every square centimetre area at the distance

of the Earth from Cen X-3 receives about 2.4 times

10 to the minus 9 ergs of x-rays

each second. If Cen X-3 radiates isotropically,

which is a highfalutin word for uniformly in all directions,

that means we can take this number and multiply it by 4

pi r squared to find out the luminosity of Cen

X-3, where r in this case is 20,000

light years, right? So, basically, what's happening

is here we are near the Earth. Here's Cen

X-3, and Cen X-3 is putting out light

12:29

You can see that each second the light from Cen

X-3 will fill up a ball whose surface area

is 4 pi d squared. So all we

have to do is take the flux, which represents one

square centimeter of area and

multiply it by all of these other square centimeters

of which there are 4 pi r squared of them, where our satellite isn't.

But which if we did have something to detect Cen X-3, we

would see the same thing as we do here near the Earth.

13:31

And if we do that, d is 20,000

light years. We multiply our flux

by the centimeter

equivalent of 20,000 light years, and if you do that, converting

light years to centimeters, you get the luminosity

of Cen X-3 is about 1.3

times 10 to the 37 ergs per second.

This is about 3000 times the entire

energy output of the Sun. And, all this from an object

whose radius is no bigger than half the length of Manhattan Island.

14:29

So we've come full circle.

It appears that the precisely varying x rays we see in Cen x-3,

GK Per and other compact x-ray binaries

are rotation hotspots associated with the star's

magnetic field. Over the years we have discovered many

such sources, each with a unique set of orbital circumstances, and each with its

own characteristic idiosyncrasies. Indeed, we have just scratched the surface

of the wonderful world of x-ray binaries. And many

surprises were in store for astronomers over the past 40

years, and many more undoubtedly will come in the future.

But now it is time to move on, and explore an entirely different class

of objects. So stay tuned as we examine our

cosmic recycling centers, Supernova

remnants, which are the products of the

most dramatic and energetic explosions

that our universe has to offer.