[mesa-users] Eddington Limit

Josiah Schwab jwschwab at berkeley.edu
Mon Dec 7 15:16:44 EST 2015

Hi Kenny,

> I was going through the code to try and set a different maximum mdot_edd
> and I noticed inside of binary_mdot.f90 that mdot_edd seems to be defined
> oddly
> if (.not. b% use_es_opacity_for_mdot_edd) then
>             mdot_edd = 4*pi*b% s1% cgrav(1)*b% m(b% a_i)/(clight*b% s1%
> opacity(1))
>          else
>             mdot_edd = 4*pi*b% s1% cgrav(1)*b% m(b% a_i)/(clight*0.2*(1+b%
> s1% surface_h1))
> Which seem to correspond to:
> Mdot = 4 pi G M / (c K)
> and
> Mdot = 4 pi G M / (c (1 + X))
> Is this equivalent to the normal representation of
> Mdot = 4 pi c R / K?
> While the units appear to match, I would assume that the two are not
> necessarily equivalent. Unless I'm misunderstanding the code here. Does
> anyone have some insight to this?

Well, it is equivalent if you set R = GM/c^2. :)

There's not a universally useful definition of Mdot_edd, given that MESA
can evolve both point mass and non-point mass companions.

Users should make use of the other_mdot_edd hook and define the Mdot_edd
that is appropriate for your problem.  

You may be interested to look at Sterl Phinney's labs from the most
recent MESA summer school.



More information about the Mesa-users mailing list