<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
<title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
Hi Massimo,<br>
<br>
I have been using FLASH's relativistic solver (RHD unit, Mignone et al
2005) to simulate astrophysical jets. Initially I did some tests with
the ppm split solver too. I didn't experience problems of instability,
except when I tried using an outflow boundary rather than a reflective
boundary on the side of the 3D box at which the jet is injected (the
outflow condition was for all rows except the jet inlet of course). I
suppose you're using a reflective boundary for the side of the 3D box
where the jet is injected? My jet parameters were constant in time. The
largest density contrast between jet and ambient medium i tried was
several 10^-5 and it was stable. I haven't tried the unsplit solver,
but I i'll let you know if I get a chance to do so.<br>
<br>
I added support for mass scalars for the RHD unit, and it was a matter
of emulating the density advection through the entire Godunov scheme.
Tedious but it in the end it seemed to work. Don't know how difficult
it would be to include mass scalars in the unsplit hydro solver, if
they aren't already supported.<br>
<br>
Sorry I can't provide more answers.<br>
<br>
Best wishes,<br>
Alex<br>
<br>
<br>
On 21/04/10 9:38 PM, Massimo Gaspari wrote:
<blockquote cite="mid:COL106-W362891EB67C04D3BFE99F9BE090@phx.gbl"
type="cite">
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Verdana
}
--></style>Dear
Flash-Users,<br>
<br>
I'm currently injecting a jet through a few cell boundaries on one side
of a 3D box. <br>
The jet has variable velocity, regulated by some near
density-temperature condition. <br>
And here come some problems with the split hydro (ppm).<br>
<br>
1) If I let vary the jet every timestep, when the velocity is
increasing the sim crashes, because<br>
the timestep (CFL condition) is fixed for 2 consecutive dt. To
circumvent this I let vary vjet<br>
only in the "second" dt (essentially fixing it also for the next one).<br>
First question: is there another (clever) simple method to avoid this
and let vjet vary every dt?<br>
<br>
2) When vjet reaches 10^4 km/sec and also when I increase resolution,
the sim easily becomes<br>
unstable at some dt (under smallest min values possible). I have to
lower CFL number to 0.1 and it goes<br>
better, but it can in any case crash in longer evolutions. Now, am I
doing something wrong with jet<br>
injection (and above set up)? 0.1 and lower make my sims almost
impossible to evolve for some Gyr, is<br>
it really possible that with ppm strang split method I have to adjust
CFL to 0.1 or lower?<br>
Again, is there any clever way to avoid this? (I know someone else has
done jets with FLASH).<br>
<br>
3) To try another way I've begun to use unsplit hydro solvers. This way
the jet can be regulated<br>
every timestep. What kind of solver do you suggest for Jet ignition (I
have not much knowledge of them)? <br>
I'm using HLLC and with 0.3 CFL the sims never crashes, also with high
vel and resolution.<br>
How can it be possible such a difference between split and unsplit
method?<br>
<br>
4) How much are the unsplit solvers in FLASH "reliable"?<br>
Have you done some tests or are they still in a "beta" phase?<br>
Again, are they adequate to study a jet ignition? <br>
<br>
5) Last point. I need also the evolution of a MASS_SCALAR and it seems
that it is not<br>
evolving with the flow with the unsplit solvers (HLLC). How can I use
them? (In Flash.h it is present)<br>
Is it possible? If not, is it easy to implement (or someone else has
done this)?<br>
<br>
Thank you very much in advance for any help.<br>
In literature I've seen many flash-users implementing jets; hope that
someone has interesting way to attach<br>
the problems I mentioned (maybe yet solved), or also completely
different approaches for jets.<br>
<br>
<br>
<br>
Max<br>
<br>
<br>
(Massimo Gaspari, University of Bologna, ITA)<br>
<br>
<hr>Novità dai Social Network e i tuoi account di posta... <a
moz-do-not-send="true"
href="http://www.windowslive.it/hotmail/UnicaCasellaDiPosta.aspx"
target="_new">Tutto su Hotmail!</a></blockquote>
<br>
</body>
</html>