Are there any limitations to passing a pPar() to a behavioural source?

Hi,

I am using a behavioural source bsource in the form of nonlinear capacitor in one of my subcircuits and I've noticed that it does not seem to recognize the pPar() construct.

If I have an analogLib/res instance, whose resistor value I declare to be pPar("myR"), then after the CDF of my subcircuit has been updated accordingly, the subcircuit is correctly showing a line

parameters myR

...

R0 resistor x y r=myR

As I would expect. I noticed however, that the same steps do not work for bsources. pPar() constructs do not give rise to a parameter declaration (once the necessary CDF adjustments have been done), but the "pPar()" string appears literally in the component value definition.

I think I've traced this behaviour back to the fact that the expression settings in the CDF are "parse as CEL=no" and "parse as number=no".

The analogLib/res, for instance has those two set to yes and this is also what I normally do if I want to be free to either set a number or a top-level variable for the expression captured by the pPar().

I tried to change the user layer of the CDF, to no avail.

I can easily give up and resort to direct variable insertion, however maybe it would be interesting to know why the difference with other components and if it can be overcome.

Thanks!

Michele