New magnetic field configurations (work in progress)

[Cham]

I'm still amazed each time I render a new field line familly. I just began the cas of inclination alpha = 45?°. The curves shape is dependant on alpha, as expected, but I also noticed it's strongly dependant on the angular velocity. At first, I believed there was a scale invariance but I was wrong on this. The magnetic field of the rotating dipole is extremelly rich and strongly depends on alpha (inclination angle) AND omega (angular velocity), so I guess I'm in trouble ! . Apparently, the case alpha = 90?° is the dullest !

Here are some pictures of a rotating dipole with alpha = 45?° and omega = 0.056 sec (again, click on the small image for a larger view) :





Of course, the green jets are wrong here, since the trapped particles are supposed to rotate around the magnetic field lines, and they can't follow them outside the "light cylinder". I'll probably remove completely the jets or replace them by a simple "schematic" green line (like the one used for the rotation axis).

[ANDREA]

Cham, the results shown in your last two posts are extraordinary, IMHO.
I have only the same little suggestion I gave you for the previous magnetic field cmod: the colors and the lines width.
I prefer, as usual, a dimmer green color and slimmer lines (the slimmest possible).
But I suppose that, as per your previous magnetic field, both values can be modified in cmod file, isn't it?
Thanks a lot, way to go!

Bye

Andrea

[t00fri]

While Cham's magnetic field lines form pleasing graphical patterns, it
seems that few users are actively interested in the crucial question
whether the physics underlying these patterns does make sense
for such "horrendous" objects like pulsars or even black holes...


Bye Fridger

[ElChristou]

While Cham's magnetic field lines form pleasing graphical patterns, it seems that few users are actively interested in the crucial question whether the physics underlying these patterns does make sense for such "horrendous" objects like pulsars or even black holes...

Bye Fridger

I suppose a closest study of those physics will be necessary IF the dev team agree in doing some test about BH and Pulsars... BTW, beside receiving the dev list and reading the forum daily, I have no idea of what's going on at dev level... any idea?

[ajtribick]

You'll like this one. Just how many magnetic poles do these pulsar things have anyway?

[t00fri]

You'll like this one. Just how many magnetic poles do these pulsar things have anyway?

The new observational evidence for a deviation from a magnetic dipole
structure does not surprise me AT ALL. I quote from the above
reference:

Tim Hankins, acting director of the Arecibo Observatory in Puerto Rico,
explained: "We think we've got a much more complicated magnetic
field than the simple dipole model.

Here is what I wrote to Cham some time ago in a PM:

honestly, I am not convinced about the physical soundness of these
studies, by Arendt and Eilek in 1998. Actually after searching ApJ it
seems that preprint was never published.

The simple modelling of some pulsar observations via a purely
electromagnetic (tilted) dipole setup, modified by rotation and
relativity, may be useful to some extent, but is not really satisfactory
to me.

Bye Fridger

[Cham]

Fridger,

of course, we all know nature's objects are always more complicated than the models we, humans, can do about them. However, the SIMPLEST pulsar model is a dipolar one. Real pulsars most probably have multipolar fields, but they surely all have a DOMINANT dipolar structure, since they obviously behave like a lighthouse. My models are just about that, with some corrections (deformation of the field from rotation).

Unless I get the right equations, that's the best I can do, especially considering Celestia's limitations on the time evolution of such objects.

And there are cases where it's much better to have an *approximate* and *crude* representation than having nothing at all.

Also, virtualy ALL pulsars representations in books, magazine, on the internet, and even in the scientific literature (almost ALL papers, articles, ...) are using a dipolar approximation. I think it could be a good thing that Celestia shows the same, UNTIL we get better.

[t00fri]

Fridger,

of course, we all know nature's objects are always more complicated
than the models we, humans, can do about them. However, the
SIMPLEST pulsar model is a dipolar one. Real pulsars most probably
have multipolar fields, but they surely all have a DOMINANT dipolar
structure, since they obviously behave like a lighthouse. My models
are just about that, with some corrections (deformation of the field
from rotation).

Unless I get the right equations, that's the best I can do, especially
considering Celestia's limitations on the time evolution of such objects.

And there are cases where it's much better to have an *approximate*
and *crude* representation than having nothing at all.

Cham,

as a theoretical physicist I am certainly ALWAYS adhering to the
strategy of successive approximations. I certainly agree about a dipole
being the simplest ansatz. But in my opinion an
electromagnetic dipole is so much oversimplified that it can hardly be
seen as a physically sensible approximation in a pulsar or even black
hole environment.

In my view, it is at least as important NOT to create the impression
among non-experts that we have a kind of quantitative understanding
of these magnetic phenomena! Your pleasing graphical display of
magnetic field lines may well support such incorrect ideas.


While I find the phenomena you obtain potentially intriguing, it is even
more crucial to FIRST be convinced about whether all this has a
chance of making sense...

Let me just indicate that your attempt of ignoring particle physics
entirely besides classical electromagnetism is very hard to understand.

In case of pulsars, we are dealing with incredibly strong
magnetic fields, heating up the surface to many millions of
degrees! Take magnetars, for example that are believed to trigger
special kinds of gamma ray bursts via "starquakes" (SGR's=soft
gamma ray repeaters etc) .


There are many sensible ideas that neutron stars could well be
(strange) quark stars that underwent a phase
transition. The magnetic field properties of such a "soup"
would be strongly modified, hardly reminding of a classical dipole field.

It just appears bold to me, to study and believe fine
details of simple rotating electromagnetic dipole fields for such
extreme environments.

Bye Fridger

[Cham]

Fridger,

let me recall you that the dipolar field is the first term (generally dominant) in any multipole expension serie for the magnetic field, whatever the source (be it quarks or other stuff). It is highly believed that the particles inside the pulsar are in some superconducting state, and should produce that dipolar field, plus (most probably) some higher multipolar fields.

Of course, I'm also interested in modeling the effects of the plasma outside the pulsar. That plasma is very likely to "drag" the field lines and make them open to the interstellar medium. This effect would be MAJOR if, in addition, there was an accretion disk around the pulsar, so the dipolar approximation doesn't make sense anymore in this case. However, modeling this for Celestia is a real challenge.

And about your concerns for the public credulity, I recall that this is just an addon project, and isn't intended for the base distribution of Celestia. In my addon, I intend to put some documentation files (all the equations used in PDF form, some references, even probably the Mathematica notebook I made) and some WARNING about the interpretation of the addon (in the readme.txt file). So I don't think there's matter for concerns here.

Here's a link to a movie showing a very complex simulation (on a supercomputer) of a realistic model of pulsar (sorry, I lost the origin of that movie, but you can trust me on this, isn't ?) :

http://nho.ohn.free.fr/celestia/Cham/Divers/t300.mov

As everybody can see, the dipolar field isn't very apparent, in this modelisation. I would love to have these field lines too in Celestia.

EDIT : I also want to make clear that what I'm doing is just a "vast" experimentation on field lines and CMOD for Celestia. The final addon (if there is really such a thing) will most probably be very different than what I'm showing right now.

Also, it's a great opportunity for me to study a fascinating physical problem that was never shown in my university studies : rotating dipole solution to the Maxwell equations. This makes the project a valuable one, in its own right, even if it never ends in an addon for Celestia.

[t00fri]

Cham,

here is another argument. Every physics student is taught that one
may expand an arbitrary electromagnetic field distribution into a
so-called multipole series. In absense of magnetic
point charges (monopoles) , such an expansion starts with the
magnetic dipole field. Next come quadrupoles etc.


You simply truncate that series without presenting the slightest
physics argument besides simplicity for truncating the series after it's
first term! In order to be able to do so, one needs to isolate a
small parameter in this expansion. In case of weak
magnetic fields such a small parameter is readily located. But not in
case of pulsars where the magnetic fields are estimated to be
one thousand trillion times the strength of Earth's
magnetic field.


Bye Fridger

EDIT: I see we were simultaneously dwelling on multipole expansions, yet with quite different implications

[Cham]

small parameter in this expansion. In case of weak
magnetic fields such a small parameter is readily located. But not in
case of pulsars where the magnetic fields are estimated to be
one thousand trillion times the strength of Earth's
magnetic field.

That's not it. The magnetic field intensity has nothing to do with the series. It has to do with the source current distribution. In the case of the UNIFORMLY magnetised SPHERE, all other expension terms CANCELS, whatever the field intensity ! For my models, I'm considering that case alone.

I never claimed my models were realist. I certainly know they aren't ! I'm just showing what the rotating dipole field may look like. It isn't a trivial field at all, so it merits that we consider that case in its own right.

[t00fri]

small parameter in this expansion. In case of weak
magnetic fields such a small parameter is readily located. But not in
case of pulsars where the magnetic fields are estimated to be
one thousand trillion times the strength of Earth's
magnetic field.

That's not it. The magnetic field intensity has nothing to do with the series. It has to do with the source current distribution. In the case of the UNIFORMLY magnetised SPHERE, all other expension terms CANCELS, whatever the field intensity ! For my models, I'm considering that case alone.

You misunderstood. I was not claiming that in strong magnetic fields a dipole structure (i.e. a cancellation of all higher terms) is IMPOSSIBLE. I merely argued that we HAVE NO HANDLE whatsoever in this case. Anything is equally possible, unlike e.g. the weak earth magnetic field, for example.

Bye Fridger

[t00fri]

...
I'm just showing what the rotating dipole field may look like. It isn't a trivial field at all, so it merits that we consider that case in its own right.

That aspect I am FULLY subscribing to and equally to the usefulness of rendering studies via CMODS of field lines in general.

Probably this was you main motivation anyhow?

Bye Fridger

[Cham]

Probably this was you main motivation anyhow?

Yes, absolutly.

Of course, I'm extremelly interested in realistic models for pulsars in Celestia. But this is (?) a project for the future, especially considering Celestia's limitations for the moment.

Also, notice that this topic's title isn't about pulsars at all

[t00fri]

Probably this was you main motivation anyhow?

Yes, absolutly.

Of course, I'm extremelly interested in realistic models for pulsars in Celestia. But this is (?) a project for the future, especially considering Celestia's limitations for the moment.

Also, notice that this topic's title isn't about pulsars at all

OK, that exchange of views was certainly useful. I was about to ask you for the actual equations that you solved, when your detailed PM arrived! Thanks, that's certainly interesting enough. I'll look at your notes as soon as I find a little time (with a fresh brain...).

Some final comments about some physics issues concerning (strange) quark stars in a superconductive phase.

The reason why superconductivity provides such a masssive effect on the magnetic/electric field is certainly familiar to you. You may also know that the color(quark) confinement in particle physics is also understood via a superconductivity picture (<-- Nobel prize winner, G. 't Hooft)! In color confinement only the role of electric and magnetic field components is flipped relative to the classical effect. In any case, in superconductivity, the magnetic field lines, say, are squeezed into thin vortices (!), on account of the Cooper pairs filling the vacuum.

Meaning a TOTAL deviation from a naive magnetic field distribution. That's why I was saying above that particle physics and its phase structure is crucial for understanding the magnetic fields around pulsars.

Incidentally, in particle physics (i.e. color-confinement) the color-electric field is squeezed into vortices and the role of the classical Cooper pairs is played by magnetic Monopole fields that are known to appear in the infrared regime of Quantum Chromodynamics (QCD) as light, dynamical degrees of freedom.

Bye Fridger

[Cham]

I prefer, as usual, a dimmer green color and slimmer lines (the slimmest possible).
But I suppose that, as per your previous magnetic field, both values can be modified in cmod file, isn't it?

Yes, the colors and transparency effect could be edited easily in the CMOD files. However, I may publish the models in binary format, because the original text version are really pretty big. The files would be MUCH smaller if I convert them to binary. I think they could be converted back to TEXT using the CMOD tool, but I never tried it for the reverse conversion.

[ANDREA]

I prefer, as usual, a dimmer green color and slimmer lines (the slimmest possible).
But I suppose that, as per your previous magnetic field, both values can be modified in cmod file, isn't it?

.... I think they could be converted back to TEXT using the CMOD tool, but I never tried it for the reverse conversion.

Yes Cham, it works.
I converted your "old" magnetic field cmod to txt, modified colors, then converted again to cmod, using cmodtools.
FYI.
Bye

Andrea

[Cham]

Just an update :

I'm getting fancier in the CMOD curves drawing. Here's the alpha = 30?° dipole, with field lines similar to what I have shown in some previous messages. I found an easy way to add a nice smooth fade out transition without seeing a sharp cut at the end of the open curves. The JPEG pictures doesn't do justice to the transition smoothness :





Here are some close view of that object, showing its evolution in time (large pictures linked) :



I recall that all the lines shown here are magnetic field lines of a real exact solution to Maxwell's equations. The purple lines are part of the radiation field.

[buggs_moran]

Wow Cham, that's amazing. The transition is quite special. Is it only viable on cmod lines? How did you do it?

[Cham]

How did you do it?

I asked Mathematica to draw the curves in several parts. After exporting all the parts to the CMOD format, I simply assigned a different material attribute to each part, with a transparency effect getting stronger to the end parts of the curves. I used 8 materials definitions in the CMOD file for the fade out transition. It's really very simple.