Typical impedance of germanium diodes?

What is a typical value range for the impedance of a germanium diode?
Say OA90 for example. HF frequencies (1-30MHz) considered.

Diode impedance depends on the circuit where it is used.
If used as a detector, it depends on input RF power. If used in a mixer, then it depends on the LO input power, and also if the mixer is balanced, etc.

Check textbooks on detectors and mixers for details.

In my circuit it is used as a doubler rectifier (2 diodes).
I would like to incluse an input attenuator pad 50R in to whatever diode impedance out, that is why I am asking.

The "impedance" of a diode is the sum of bulk resistance and differential resistance rd = dV/dI which is current dependent according to the exponential diode characteristic.

You can derive both numbers from the datasheet forward characteristic. OA 90 bulk resistance range is 25 - 100 ohm according to datasheet, typically 50 ohm.

So it is roughly already matched or I have to add the Rd to this value as well?

The circuit I consider is this. It uses a aproximately 50R input attenuator to the diodes. The diodes are used as charge pumps, so I think impedance matching would not be too critical on HF anyway.

In this app, Rd is varied so I think I cannot consider constant impedance...

For such application I am not sure what input power is. I used GaAs Schottky diode pairs as doublers, and under 17-18 dBm input the input impedance was close to 50 Ohms, also over a wide band.

The best way to know is to use a reflectometer and measure S11 as a function of input power. Use such level that gives the best ouput, with luck this may coincide with a good match.

You need a capacitor between the 500 ohms and the two diodes, else the diode connected to earth will short out the RF.
Frank

But isnt't that the case even with the capacitor in place?
During the negative half of the sinewave, the shunt diode conducts rf to the GND.



Never mind, you are right, it is a voltage doubler configuration, it needs a capacitor at the input.
Thanks for pointing it out