How to calculate parasitic capacitance from |Z|/R graph

[Palpurul]

Hello, I've been trying to calculate the parasitic capacitance(end-to-end not pad to ground) of resistors from |z|/R graph that is given in some of the resistors' datasheets. Here is how I calculate it. I use the following circuit to model the resistor. I measure the input impedance of this circuit to ground. This model is also used in the datasheets( datasheets are shared below).



When I simulated this I discovered that at approximately |Z|/R = 0.7 frequency is f= 1/(2*pi*R*C). In this document http://www.farnell.com/datasheets/6370.pdf for internal impedance (end-to-end) for 0402 flip chip is given by 26.2 fF with that much capacitance according to my calculation for 1k resistor the |Z|/R graph should hit 0.7 at around 6 GHz but in the datasheet 0.7 crossing is between 2-3 GHz. What am I doing wrong here?

I am trying to calculate the capacitance of the resistors in this document : **broken link removed**(Fig.4 0402 components)

I calculate around 31fF (0.7 crossing around 50Mhz for 100k applying C=1/(2*pi*R*f)) , but I am skeptic about this because I am not sure if I am doing it right.

Thanks in advance.

 

[Palpurul]

Does anyone have any idea about SMD resistors parasitic capacitance in general? I am going to be using 0402 thick film resistors from yageo. **broken link removed**
In this document |Z|/R graph is given but the value of the parasitic capacitance is not given. How can I extract the value of parasitic capacitance from this graphs for 0402 components(fig.4)?

 

[FvM]

The correct link is in post #1, don't copy and past abbreviated links.

I see that the model data in the Vishay data sheet are apparently copied from a more detailed application note https://www.vishay.com/docs/60107/freqresp.pdf

I understand that the "C" values are assigned to the naked chip while the impedance curves include the external impedances, particularly Cg. This fact seems to be partly confused in the datasheet quote of the measurements, but it's quite obvious that real measurements need some kind of test fixture. As you'll probably use FR4 or RF substrates other than quartz, you have to put in layout impedances of your actual hardware.

I believe however that resistors with similar shape (e.g. industry standard 0402 "wrap around" chips) will expose similar capacitance values. Inductance might vary depending on trimming method.