I'm currently trying to build my own SWR power Meter. This design is based on the so known HT detector with schottky diode BAS70. For my understanding, and for easy purpose, I'm currently focusing on 50 MHz.
First, I found a nice theory called "The Square Law Region" which can be found here on page 115 of the book Radio Frequency & Microwave power measurement.
It says that voltage out of the HF detector is $$V=\frac{25e}{kTn}P_{RF}$$
where $e$ is the charge of an electron ($-1.602 \times 10^{-19}$ C), $k$ is the Boltzmann constant $1.38064852 \times 10^{-23} m^2 kg \cdot s^{-2} K^{-1}$, $T$ the temperature in degrees Kelvin, $n$ the ideality factor, and $P_{RF}$ the power in Watts.
I found that Spice says $n=1.018$ for BAS70 diode. If my calculation is correct, for P = -30dBm (1µW) lead to out voltage to 950µV. Is that correct?
At 50MHz, with BAS70, 47nF, it tried this set up...my measure leads to about 10mV...wich is far away for...let's say 1mV of theory. Is this theory correct ?
Then I also tried to set up in simulation with Qucs. I faced another problem: Qucs failed to simulate this situation.
First, I must say that I see different delay of response regarding the input level. For +10dBm, voltage out tends to 1.5V in about 300µS. For -30dBm, voltage is not stable before 1mS (see jpg joined file). I understand that it could come from the time needed by capacitor...but also, timing response should not depend on the input level for a basic RC filter. Right?
What should I do to simulate my HF detector at low level?