All,
Is it true that the particletemplate/sphere command actually has styles other than "constant" for the density and radius keywords? I am using a combination of 3.3.1 and 3.1.0 (the older one is on my school's HPC cluster, and I don't have the knowledge/rights to upgrade it), and I am still using the "uniform" style for radius (i.e. ...radius uniform number rmin rmax). I am simulating "real" particles, which are often given a size distribution according to mesh size, and while the uniform keyword isn't really how the actual PSD looks, it is a very convenient way to get approximate mass % in the appropriate mesh bins for whatever material I'm using.
Just thought I would point this out. The current manual states that only the "constant" style will work, but my dumpfiles seem to show a uniform distribution of particle sizes per particletemplate/sphere. For me personally, this is a good feature, and should be left in! I believe there used to also be a "gaussian" option, which would also be handy to get a slightly more realistic PSD. I understand that real particles follow a Weibull distribution, and can often use a log-normal distribution which is a bit mathematically simpler, but those would still be excellent features to put in (or officially reinstate in the manual).
FYI
ckloss | Mon, 01/02/2017 - 14:01
Hi msandli,
Hi msandli,
at the moment only "constant" is implemented, and any arbitrary distribution can be mimicked by a combination of those. I understand that it's useful to put in a distribution function, but it would have to be re-implemented in a different way than it was implemented in earlier versions - this was discontinued for efficiency reasons
Best wishes
Christoph