Hello everyone!

For some context on my question, I am working on railway vehicle simulation and I must create track irregularities to better model a realistic railway. So my question is more to ask for someone to verify if the procedure my logic is correct. I will also highlight my more specific doubts. 

First, I must create a Power Spectral Density function (PSD - stochastic 108) from a formula given in the first attached image. This image is for an FRA vertical track irregularity. As seen in the image, a polinomial quocient can describe this formula and SIMPACK provides this feature (second attached image).

The problem here are the units of the adopted formula. By plugging in the units into the formula, the output would be in cm^2/(rad/m). Comparing this to the example of a predifined PSD in SIMPACK (ERRI B176), their units are m^2/(rad/m). 

Question 1: I have to include a factor of 10^-4 to obtain a spectrum in meters instead of centimeters since I would go from cm^2/(rad/m) to m^2/(rad/m) correct?

Continuing the process, I must define the lower and upper bound of the irregularity spectrum for creating the track excitation from type 108. Since the PSD is now in m^2/(rad/m), my intuition leads me to believe that the upper and lower bound must have the same units as the denominator of the PSD (rad/m). The hover tooltip, however, states that the upper frequency limit and lower frequency limit must have units (1/m), however, I believe that it should be (rad/m) given the definition of the spectrum. 

Question 2: Does the spectrum have to be in m^2/(rad/m) and the frequency limits in (1/m) to correctly obtain the irregularities in meters? Consider now that the spectrum is now m^2/(1/m) because the angular frequency "w=2*pi*f" was replaced by frequency "f". Is the definition of the frequency limits in (1/m) still correct?

I know this is a very long question but I would really appreciate the feeback. Thank you for your time!