Hi Miguel,

Here's a bit more information. There is no support in McLachlan for using LegoExcision. LegoExcision was used with an old implementation of BSSN. As Jonah said, when using Llama, it's possible to do excision by using a six-patch coordinate setup. In principle it is also possible to use a patch system with 13 patches to have 2 spherical exicision regions. However, there is no infrastructure for keeping 2 black holes centered on the excision regions as the system starts to evolve. Hence, you'll encounter the problem that the black holes will move and the excision regions will eventually cross the horizon at which point your simulation will crash.

To perform simulations a lot of infrastructure (a la what is available in SpEC) to control the shape and size of the horizons during a dynamical evolution.

Cheers,

Peter

On Monday 2017-03-27 09:59, Jonah Miller wrote:

Date: Mon, 27 Mar 2017 09:59:51 From: Jonah Miller jonah.maxwell.miller@gmail.com To: users@einsteintoolkit.org Subject: Re: [Users] Excision with McLachlan

Hi Miguel,

I'm not sure about excising dynamically during evolution--maybe someone else has insight there. However, excision is definitely possible. I have been using the sixpatches Llama thorn to perform excision for a single black hole. I attach a parameter file to excise a Kerr-Schild black hole that I based my simulations on.

Best,

Jonah

On 2017-03-27 05:01 AM, Miguel Zilhão wrote:

...

hi all,

i was wondering whether one could straightforwardly use the ET to perform black hole evolutions using excision instead of punctures.

specifically, i had in mind a collapsing configuration where i'd like to excise the inner domain once a black hole horizon is formed... is this possible? i had a look around and noticed the LegoExcision thorn, but i'm not sure whether it's actively maintained, nor whether it can interface with McLachlan...

thanks, Miguel _______________________________________________ Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users

On Mon, Mar 27, 2017 at 11:28:42AM -0500, Frank Loeffler wrote:

it seems "good enough" to apply a little more dissipation well within the horizon to not crash - and that is really all you need. Nothing bad seems to "come out" of the black hole. You simply evolve everywhere and

Just to clarify: You of course need to check that this is indeed the case in your simulations. What we found is that it is possible to do that for the cases we looked at, and it wasn't especially difficult or finicky then.

Frank

Hi Miguel,

1. For the geometry, I believe you can choose between extrapolated and radiative boundary conditions. I think the excision boundary has the same conditions as the outer boundary, which are set in the parameter file as:

ML_BSSN::initial_boundary_condition = "extrapolate-gammas" ML_BSSN::rhs_boundary_condition = "NewRad" Boundary::radpower = 2

Others can correct me if this is not the case. For matter, outflow boundary conditions are probably appropriate. Usually it doesn't matter what boundary conditions you set at the excision boundary, since nothing will escape the horizon anyway.

2. I don't think it's possible to excise dynamically... at least not with Llama. The way this infrastructure works is that you create a grid made out of 6 cubed spheres. So the excision is actually built into the coordinate system--it's the inner boundary of each cubed sphere.

You could get around this by, for example, stopping the simulation when the apparent horizon appears and then interpolating onto the sixpatches grid and restarting the simulation.

Or you could follow Frank's suggestion and, rather than excising, simply under-resolve/dissipate away whatever is going on inside the black hole.

Best,

Jonah

On 17-03-28 07:05 PM, Miguel Zilhão wrote:

hi all,

many thanks for the many replies, in particular to Jonah for providing the parameter file! i have just a couple of follow up questions:

• what sort of boundary conditions are imposed on the evolved

variables inside the excised region, and how is this controlled?

• would it then be possible to do this dynamically, ie, start

excising only when a certain condition (like formation of an apparent horizon) is fulfilled?

and for my purposes having a single excised region is fine, the setup i have in mind would be a collapse to a single black hole.

thanks, Miguel

On 27/03/17 16:59, Jonah Miller wrote:

...

Hi Miguel,

I'm not sure about excising dynamically during evolution--maybe someone else has insight there. However, excision is definitely possible. I have been using the sixpatches Llama thorn to perform excision for a single black hole. I attach a parameter file to excise a Kerr-Schild black hole that I based my simulations on.

Best,

Jonah

On 2017-03-27 05:01 AM, Miguel Zilhão wrote:

...

hi all,

i was wondering whether one could straightforwardly use the ET to perform black hole evolutions using excision instead of punctures.

specifically, i had in mind a collapsing configuration where i'd like to excise the inner domain once a black hole horizon is formed... is this possible? i had a look around and noticed the LegoExcision thorn, but i'm not sure whether it's actively maintained, nor whether it can interface with McLachlan...

thanks, Miguel _______________________________________________ Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users

---

Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users