Looking at Planck units, there seems to be a curious rule between the dependance in $\hbar$ of a Planck unit and the unit dimensions of the corresponding physical quantity.
Let the dimensions of the physical quantity be ($Q$ being the electric charge unit and $\Theta$ being the temperature unit):
$$ L^l M^m T^t Q^q \Theta^\theta.$$
Then, if : $$l + m + t + q + \theta = 0$$
the Planck unit does not depend on $\hbar$.
This seems to work for all base Planck units, and, consequently, for all derived Planck units.
Is it just chance, or is there a more fundamental reason?