[FLASH-USERS] Unexpected instable shell interface in cylindrical implosion simulation

[Mathieu Bailly-Grandvaux]

Dear All,

Thank you for your responses and helpful insights!
@rjfarber at umich.edu <rjfarber at umich.edu>: Yes, I believe we can rule out
grid artifacts due to resolution. I do expect RT growth as a physical
phenomenon; however, the observed level of RT growth appears unphysically
high when compared to both GORGON 2D simulations and experimental data. The
shell seems to get "punctured" early (see at t = 2.5 ns it almost made a
hole through its 50 um thickness), likely due to laser imprint artifacts.
These artifacts are subsequently strongly amplified by RT growth, leading
to what appears to be artificial and exaggerated shell shredding. Regarding
the "ForceHydroLimit = .true." option, I recall trying this previously
without observing significant changes, but I didn’t save that run. I will
revisit this to confirm whether the MHD model could be contributing to the
issue and will let you know the outcome.

@fernando at pacificfusion.com <fernando at pacificfusion.com>: Exactly—this
aligns with my thoughts as well, as discussed above. I previously attempted
to include a pre-heating beam to allow the shell to expand slightly before
driving it, hoping that the resulting heated gradient scale length would
mitigate the artificial RT growth. Unfortunately, this approach was not
particularly successful. However, I haven’t yet tried directly modifying
the initial density profile, and my "vacuum" is still set at room
temperature. Thank you for this suggestion—I’ll explore that approach!

@Joshi, Tirtha <tjoshi at lle.rochester.edu>: When I employed the 3D ray
tracing, I used the "statistical2D" configuration. However, for this case,
since I am using 2D irradiation, I’ve switched to "statistical1D".

Thank you again for your input, and I’ll follow up with updates as I test
these suggestions.

Best,
Mathieu

On Fri, Dec 20, 2024 at 1:23 PM Joshi, Tirtha <tjoshi at lle.rochester.edu>
wrote:

> Hi Fernando and Mathieu,
>
> I had also observed similar effects in my laser-slab simulations, and
> these effects were reduced when I replaced the ed_gridType_1 = "radial2D"
>
> by
>
>  ed_gridType_1                 = "statistical2D".
>
>
> Best,
> Tirtha
>
>
> ------------------------------
> *From:* flash-users <flash-users-bounces at flash.rochester.edu> on behalf
> of Fernando Garcia Rubio <fernando at pacificfusion.com>
> *Sent:* Friday, December 20, 2024 3:48 PM
> *To:* Mathieu Bailly-Grandvaux <mbaillygrandvaux at ucsd.edu>
> *Cc:* flash-users at flash.rochester.edu <flash-users at flash.rochester.edu>
> *Subject:* Re: [FLASH-USERS] Unexpected instable shell interface in
> cylindrical implosion simulation
>
> Hi Mathieu,
>
> I also noticed that the laser package in FLASH does not currently have
> energy smoothing techniques, and as such might incur in significant laser
> imprint on targets. Some times ablation is enough to smooth these
> perturbations out, but for very thin targets,  this results in an
> artificial enhancement of RTI and subsequent target shred.
>
> What has worked for me in the past is to increase the “vacuum” temperature
> and impose a tanh(x) density profile for the large edge so that thermal
> conduction can naturally smooth the imprinted perturbations.
>
> Hope this helps!
>
> Fernando
>
>
> On Dec 20, 2024, at 11:04 AM, Mathieu Bailly-Grandvaux <
> mbaillygrandvaux at ucsd.edu> wrote:
>
> Dear FLASH Users,
>
> I am using FLASH to simulate a cylindrical implosion for the National
> Ignition Facility (NIF). The simulations are performed in 2D r-z geometry.
>
> Initially, I utilized the 3D-to-2D ray tracing algorithm, incorporating
> the 128 beams of the experimental configuration. I observed a rapid growth
> of instabilities at the gas-shell interface and initially considered that
> this might be a (potentially physical) result of azimuthal asymmetry
> (mode-8 is particularly expected) due to the beam imprint pattern projected
> in 2D. However, the asymmetries were far more pronounced than anticipated.
>
> To rule out this effect, I simplified the simulations by switching to a 2D
> irradiation model using an azimuthally averaged "gaussian1D" function.
> Surprisingly, the implosions remained highly unstable, indicating that the
> projected azimuthal asymmetries were not the cause this time. I
> experimented with varying the number of rays, ticks, and grid resolution,
> but the instability persisted across all configurations.
>
> At this point, I am uncertain about how to resolve this issue, and would
> greatly appreciate your advice and insights.
>
> I have attached the flash.par and setup_call files for the simulation,
> which was conducted using FLASH version 4.8. Below, I include mass density
> plots ("dens" in g/cm³) at several time steps (2.5 ns, 5 ns, and 7.5 ns out
> of the 10 ns pulse duration) to illustrate the instabilities observed:
>
> <image.png><image.png><image.png>
>
> Additionally, I enabled the MHD capability with an axial B-field, but it
> is currently set to B = 0 T in this simulation.
>
> Thank you in advance for your time and assistance. I look forward to
> hearing your thoughts and potential suggestions.
> Best regards,
> Mathieu Bailly-Grandvaux
> <flash.par><setup_call>_______________________________________________
> flash-users mailing list
> flash-users at flash.rochester.edu
>
> For list info, including unsubscribe:
> https://flash.rochester.edu/mailman/listinfo/flash-users
> <https://urldefense.com/v3/__https://flash.rochester.edu/mailman/listinfo/flash-users__;!!Mih3wA!A_Vwg2PBb8_IwkFDaR5qiAzixNBMOnPU5vhPg_QK0H92jVB7STQCcm86QJOqMpmgBttoMYi2vM39QjX_WRk84ZSIij1kvA$>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://flash.rochester.edu/pipermail/flash-users/attachments/20241223/826e5f9c/attachment.htm>