![Image](http://www.postimage.org/Pq15JfLA.jpg)
This uses the classic and often seen low junction drop of a germanium 1N34 diode. The Vf as measured on a Fluke DMM is 0.2 volts (not sure what current it injects to obtain this voltage). This gives me fairly linear responses in the Square Law region of 200 - 2000 mv P-P input and then when moving beyond the transition zone into the linear region - excellent response, but at the expense of requiring high RF level input. Before I delve into a lengthy search on low Vf diodes (and thus lower detection levels), I thoght I would pick the expertise of the forum for any knowledge of very low Vf diodes. Of course junction capacitance must be kept to a minimum (<2pf) due to the frequencies involved. I do remember HP having a diode with a Vf of 0.1 volt from years ago, but have long since lost track of it.
Also, I tried biasing this diode with a small positive voltage on its anode with some unexpected results-that being the only difference in operation was the addition of a DC Vf voltage appearing on the output leads and no change in signal detection performance. This may be due to the hi-'Z' environment in which the circuit operates in, greatly limiting the actual current thru the diode. Would like to keep ths as a simple two wire device, but I suppose I could add a third wire to it for biasing if warranted. Looking to detect 20 mv p-p RF and higher with some degree of accuracy and only need relative output values (not absolute) to the scope and temperature stability will be of little concern as it will be operated in a room temperature environment all the time.