Recreating a theoretical star system

[tuckrr]

Hey guys! 
I'm new to using universe sandbox, but so far I think its great.  I've been trying to recreate a binary star system which was theorized in 2001.  The paper, written by S. Edgeworth, can be found here:
http://www.academia.edu/3029649/Theoretical_Orbits_of_Planets_in_Binary_Star_Systems

I'm specifically trying to achieve page #15/16

I don't know enough about the program yet to make this system myself, but I'm learning quickly.  If someone knows a guru on the forums please let them know about this challenging system.  I'd really appreciate some help with the advanced tools, or getting a stable starting point.
Cheers,
Tuckrr

[gabriel.dac]

(Link of the orbit, for the lazy: http://htmlimg1.scribdassets.com/2lq25ze3pc2eifzu/images/15-7380fd2fe7.jpg)

haha wow. That's something extremely tricky to do! I'm afraid I can't help you on that

[tuckrr]

Also, a bug I've noticed with the software.  Binary star systems don't last very long.
Repro Steps:
1.  Make 2 stars of equal properties
2.  Highlight both
3.  Click "Make Binary Orbit"
4.  Click "Add Barycenter"
5.  Click "Balance selected"
6.  Speed up the time, after a few years the stars will leave each other somehow.

Weird bug, but it will make stable systems tough to achieve...

[Unnamed25]

Also, a bug I've noticed with the software.  Binary star systems don't last very long.
Repro Steps:
1.  Make 2 stars of equal properties
2.  Highlight both
3.  Click "Make Binary Orbit"
4.  Click "Add Barycenter"
5.  Click "Balance selected"
6.  Speed up the time, after a few years the stars will leave each other somehow.

Weird bug, but it will make stable systems tough to achieve...

The faster the simulation goes, the more inaccurate it gets.

[tuckrr]

The faster the simulation goes, the more inaccurate it gets.

Ahh Gotcha.  Is there a recommended speed I should use?

Also, are there templates of Binary systems I can work off of?

[sedge]

Hi, I wrote that article, thanks for reading it.
I don't have universe sandbox so am guessing here:
First set up a binary system of two stars of equal mass with circular orbits around each other.
If the orbits decay, reduce the time step (slow it down).
After getting the two stars orbiting ok, you then need to add a planet of zero(negligible) mass.
To give you an idea of where to position this planet, and what velocity to give it, here are the initial parameters for a similar orbit illustrated on page 17.
https://www.academia.edu/4716055/Initial_parameters_for_a_theoretical_S-type_stable_planetary_orbit_with_period_equal_to_the_binary_star_period

[Unnamed25]

The faster the simulation goes, the more inaccurate it gets.

Ahh Gotcha.  Is there a recommended speed I should use?

Also, are there templates of Binary systems I can work off of?

It depends on computer power.

A more powerful computer can go faster timesteps without starting to get really innacurate