[SOLVED] RF diode vs standard diode

I need to rectify frequency in the range of approx 100MHz.
There are standard schottky diodes with around 1ns reverse recovery time. These are 'ok-ish' but I want to know if RF diodes are a better alternative. I looked at a few RF diode datasheets, but they dont specify the reverse recovery time. Its just the capacitance that's mentioned.(around 2pF)
What's the difference between a schottky and a RF diode in terms of speed?
Thanks

Usually RF diodes are used with low signal voltages. One use is to exploit the non-linear current-voltage characteristics to change the RF frequency, e.g. in a mixer. Another use is to change an applied DC current so that the dynamic small-signal resistance changes, thus acting as an attenuator.

In theory, schottky diodes have zero reverse recovery time. https://en.wikipedia.org/wiki/Schottky_diode#Reverse_recovery_time

I presume you are referring with "RF diodes" to small regular Si pn diodes?

At low current and high frequency, diode behaviour will be mostly ruled by diode capacitance. I would expect small, low capacitance schottky (detector) diodes giving best rectifier performance.

FvM said:

In theory, schottky diodes have zero reverse recovery time. https://en.wikipedia.org/wiki/Schottky_diode#Reverse_recovery_time

I presume you are referring with "RF diodes" to small regular Si pn diodes?

At low current and high frequency, diode behaviour will be mostly ruled by diode capacitance. I would expect small, low capacitance schottky (detector) diodes giving best rectifier performance.

Click to expand...

To get a correct answer please indicate what you mean by "rectifiying a frequency of 100 MHz". You can rectify a SIGNAL, then the specification of power level and output load must be given. Schottky diodes come in various versions, some for low-level detection like -30 dBm, others capable to rectify half-watt of power at 100 MHz.

schottky diodes have two big advantages:
low junction voltage, so they will respond to smaller RF signals
a very thin junction region, so there are few transit time effects

but at 100 MHz, a standard silicon junction diode will probably work. And you can always partly bias the junction diode on with a few uA of forward current to help it respond to small signals.

In either case, you want a small capacitance. At 100 MHz, I would think anything below ~12 pF would do nicely.

For an RF detector the type of the rectifying component (diode, transistor, tube, etc) it is important, but more important for getting good performances is the biasing and the input matching.

In a low voltage Schottky diode detector, the optimum sensitivity occurs with bias level typically between 50-70 μA.
The maximum sensitivity occurs when the diode bias current is at a point of maximum curvature of the forward I-V characteristics.
The Schottky diode impedance can be calculated in the square law range (about -20dBm to -10dBm) by using the approximation :

Rd = (28/I)*1000 where I is the diode bias current expressed in uA.

For example a bias level of 80uA gives a Schottky detector impedance of about 350 ohms.
Knowing the input impedance of the detector, can avoid any mismatches that can appear.

In 100MHz frequency range, biased Germanium diodes (even they have higher knee voltage than Schottky diodes), can provide lower input impedance for the same bias current, which might be better for RF applications.