Dear Colleague,

A number of bugs in both the Lattice Boltzmann and Dissipative Particle Dynamics 
codes, and in the JAVA GUI, have been discovered.

Three bugs were discovered in the Lattice Boltzmann code:

1. The implemented Zou/He boundary conditions for constant velocities and
   densities cannot be used for fixed solute concentration and temperature 
   boundaries. This scheme was found to produce excessively high values for the 
   specified properties, most likely because the velocities derived from 
   separate distribution functions for solutes and thermal flows do not 
   correspond with the real fluid velocities.
2. The four routines reading lbin.init files in lbpIO.cpp read the incorrect 
   number of fluid densities (pressures) for the number of fluids being 
   simulated.
3. The utility lbeplot3dgather cannot read nor write two-dimensional grid files.

One bug was found in the Dissipative Particle Dynamics code: the routine 
bondforcesglobal (in bond_module.f90) uses three arrays to store and share the 
locations of beads in molecular structures between all processors. These arrays, 
however, are not set equal to zero each time the routine is called. As a result 
the arrays may contain spurious values depending on the compiler used and cause 
floating point errors. (Many thanks are due to Edward Kim for bringing our 
attention to this bug.)

Two minor bugs were also discovered in the JAVA GUI: both of these caused 
problems when specifying thermal parameters and boundary conditions for two-
dimensional systems.

The solutions to the second and third bugs in the LBE code and the bugs in the 
DPD code and JAVA GUI require minor edits. The solution for the first LBE bug, 
however, has involved writing additional routines to apply the Inamuro boundary 
scheme for constant solute concentrations and temperatures as well as 
restructuring the routines for constant fluid density/velocity conditions (which 
still use the Zou/He scheme). As a result, the solutions to these bugs will not 
be given in this email.

We strongly recommend that all registered users download DL_MESO again using the 
instructions they received on registration to obtain the corrected version of 
the software. Re-registration is not required and the same password is used for 
decryption. An additional section on the revised LBE boundary conditions for 
solute concentrations and temperature can be found in the revised version of the 
user manual.

Michael Seaton
2011-06-17