I'm doing a TRL calibration for the first time and I came across Ap 1364-1 (page 12). Is that formula correct? The example it goes through doesn't seem to add up. What are the units supposed to be for that formula?

What I assume:

Offset loss= IL[dB] * c *sqrt(er) *Z0/ (10 *log(e) *L)

er=permittivity=4.3

c=3e8

Z0=50

e=exp(1)

L=3in *.0254meters/in

IL @ 4.5GHz ~ 1.25dB

Problem's offset: 10 Gohms/s

My calculation from that formula:1.12789e+11 Gohms/s

http://ece.uprm.edu/~colom/de-embedding.pdf

What I assume:

Offset loss= IL[dB] * c *sqrt(er) *Z0/ (10 *log(e) *L)

er=permittivity=4.3

c=3e8

Z0=50

e=exp(1)

L=3in *.0254meters/in

IL @ 4.5GHz ~ 1.25dB

Problem's offset: 10 Gohms/s

My calculation from that formula:1.12789e+11 Gohms/s

http://ece.uprm.edu/~colom/de-embedding.pdf

offset loss (dB) = (loss [dB at 1 GHz] * Zo) / (10 * ln(10) * offset delay)

Substituting delay = L * SQRT(er) / c, where L is the physical length in meters, er is the relative permittivity, and c is the speed of light in m/s, we get

offset loss (dB) = (loss [dB at 1 GHz] * Zo * c) / (10 * ln(10) * L * SQRT(er))