Added some explanatory comments on why the calculations sometimes result in a merge

src/symba/symba_frag_pos.f90

@@ -367,6 +367,8 @@ subroutine symba_frag_pos_kinetic_energy(xcom, vcom, L_orb, L_spin, m_frag, x_fr
            v_frag(:, i) = v_r_mag(i) * v_r_unit(:, i) + v_t(:, i)
          end do
       else
+         ! No solution exists for this case, so we need to indicate that this should be a merge
+         ! This may happen due to setting the tangential velocities too high when setting the angular momentum constraint
          lmerge = .true.
          v_frag(:, :) = 0.0_DP
       end if