Volume limit particletemplate/multisphere not working / contact problem with "micro" units?

Hello,

I have coded an algorithm to generate multi-spheres to describe the shapes I need.

The problem I have now is that, since the spheres generating the multi-spheres are in the order of microns (10-60 microns) and use units SI, when I try to run the LIGGGHTS script I get the following error:

ERROR: Fix particletemplate/multisphere (id pts1): Volume expectancy too small. Change 'volume_limit' if you are sure you know what you're doing (../fix_template_sphere.cpp:447)

Unfortunately, the particletemplate/multisphere does not have the "volume_limit" parameter as the particletemplate/sphere has so I cannot set a lower volume limit. Anyone knows if this is going to be fixed in a future release?

For now, what I would try to do is to pass from "SI" units to "micro" units, even in the particles are contained in a box in the order of millimeters (about 100mm).
Reading the LIGGGHTS documentation I am not quite sure what I should use as pressure and dynamic viscosity.

Could you please confirm the following:

to convert the pressure from Pascals (SI units) to picogram/(micrometer-microsecond^2) (micro units) I have to multiply by 10^-12
to convert the dynamic viscosity from Pascal*second (SI units) to picogram/(micrometer-microsecond) (micro units) I have to multiply by 10^9

I guess that at the denominator of pressure and dynamic viscosity, in the following link
http://www.cfdem.com/media/DEM/docu/units.html
between micrometer and microsecond there is a "*" instead of a "-"?

For now, comparing a model with "SI" units with the same model but using "micro" units, it seems that I get the same motion of the particles so, this would be a first confirmation I have converted the units correctly.
Unfortunately, in the model with "micro" units it seems that the contact does not work properly and particles do not detect the wall of the box. Is there any reason?
See file attached to understand the problem. It seems that the particles, moving toward the -Z direction, do not detect the surface. And this seems to happen when I use the "micro" units.

Thanks
Matteo