Hi Will,
Artificial compressibility factor sets the pseudo speed of sound and
therefore adjusts the characteristics of the system. Typical values for
ac-zeta are (1 -- 10)*maximum velocity magnitude. If you set ac-zeta
very large, it limits your explicit pseudo-time step and kills the
performance.
A strong pseudo pressure wave occurs always in the beginning of the
simulation. With a large ac-zeta value, it seems weaker than with lower
ac-zeta because it has travelled further with respect to the flow field
(pseudo-speed of sound/flow speed). In an ideal case, this initial wave
would be killed during the first physical time step by just performing a
lot of iterations, but it is not feasible. Therefore, the residual of
this initial wave tends to propagate in physical time during the initial
transient phase of the simulation. With external flows the wave can be
easily dissipated by a sponge region at the far-field boundaries. In
internal flows you just need let it bounce during the initial transient
phase and wait until it is naturally dissipated by viscosity. Please not
that this wave occurs only in the very beginning when you start the
simulation from initial conditions that are "far away" from the actual
solution. You only need to develop the flow once and then you can keep
restarting from the developed solution to gather statistics.
About the boundary conditions. The values in the solution file are
domain-sided values from the polynomial representation of the solution.
Remember that the solution is discontinuous in FR! The actual boundary
velocity would be
BCvelocity = 0.5*(u_domainside + u_ghostside),
which would be exactly one for the lid in the cavity flow.
Regards,
Niki
On 16/11/17 07:40, Will wrote:
Dear Niki,
Another thing.
I dont know whether my understanding is right. To approximate
incompressibility through AC method, the AC factor should be
relatively very large than pressure so that the continuity eqn would
be very close to the incompressible continuity eqn.
I tried a test in 2d internal pipe flow with constant pressure
gradient of 4. There seemed to be something like normal shock happened
when ac-factor/P = 0.5, while if ac-factor/P reached 10,000, the flow
is quite incompressible.
Does ac-factor/P matter the incompressibility of the flow?
Best regards,
Will
--
You received this message because you are subscribed to the Google
Groups "PyFR Mailing List" group.
To unsubscribe from this group and stop receiving emails from it, send
an email to [email protected]
<mailto:[email protected]>.
To post to this group, send email to [email protected]
<mailto:[email protected]>.
Visit this group at https://groups.google.com/group/pyfrmailinglist.
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google Groups "PyFR
Mailing List" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send an email to [email protected].
Visit this group at https://groups.google.com/group/pyfrmailinglist.
For more options, visit https://groups.google.com/d/optout.
