The risetime in your article is predicated on sinusoid as base signal of interest.
Digital systems largely governed by L & C, C largely. And current availability to ramp the
C. Lastly the technology of the logic, several different CMOS and bipolar families.
So no, you would simulate or look at digital outputs drive capability to determine Tr and Tf.
In addition:
even for a sine waveform it´s not "the rise time" because rise rate continously changes.
At 0° it is at positive maximum
at 90° and 270° it is zero
at 270° it is negative maximum.
--> So what the formula actually calcuates is the "maximum rise rate" for a sine.
As mentioned in the article, rise time is usually defined as time interval between 10% and 90% level of a rising edge. The parameter is e.g. automatically determined by digital oscilloscopes and can be of course also applied to sine waves, although I doubt that makes sense. The discussed 1/(3*f) or 1/(pi*f) number neither fits 10/90 nor 20/80 risetime, in my view the reference to sine waves is misleading.
The commonly used relation is between rise time and bandwidth of a measurement channel. It can be derived from first or second order transfer function and gives about tr = 1/(3*fg). My old Tek 485 has e.g. 350 MHz bandwidth and 1 ns 10/90 risetime.
Trise definition seems, appropriately, to change for system greater that first order,
which makes sense. E.g. damping factor qualified. Here is interesting derivation
and definition :
The order of a control system is determined by the power of ‘s’ in the denominator of its transfer function. If the power of s in the denominator of the transfer function of a control system is 2, then the system is said to be second order control system. The…