[Fleiga]

Hi to all,

I want to simulate a granular pile collapse and therefore inject 91 parcels containing 1 particle each, running the DDPM-model in a 2D Domain (1m x 0.5m). Everytime i inject my particles at time 0 for the time step 1e-4 the particle parcels are not uniformly sorted, as my injection file suggests. They look randomly placed. Here in short the injection file:

(( 0.4121765875 0.014611905000000001 0 0 0 0 0.024353175 298.15 31.762500247718553 ) injection0:1)
(( 0.43896508 0.014611905000000001 0 0 0 0 0.024353175 298.15 31.762500247718553 ) injection0:2)
(( 0.46575357250000005 0.014611905000000001 0 0 0 0 0.024353175 298.15 31.762500247718553 ) injection0:3)
(( 0.49254206500000003 0.014611905000000001 0 0 0 0 0.024353175 298.15 31.762500247718553 ) injection0:4)
(( 0.5193305575 0.014611905000000001 0 0 0 0 0.024353175 298.15 31.762500247718553 ) injection0:5)
...
...


Additionally, after the injection. The following message occurs:

"Injection injection-wood: Out of 91 locations, 1 lost outside the domain in radial staggering; reducing staggering factor to 80% for them.
Injecting 91 particle parcels with mass 0.289039 at t = 0"



Surprisingly, when i run my case in Fluent 2020 the injection works fine, as i expect. Also when i inject 141 particle parcels the error where i lose a particle is not showing any more. Anyway, i have to run i at Fluent 19.0 because of more computational power. Has anybody faced this problem before? I simply want my parcels look like in the file. Changing the injection type and initilaze first, before the actual simulation hast not helped me at all. The particles are not overlaping or outside the domain, i guess (d_p = 0.024353175 m).

Cheers Fleiga

[DLL123]

did you try to reduce your stagger radius?