HF band (500KHz to 30MHz)

I'm in need of an input protection circuit for the front-end of an HF receiver. I'm currently using Schottky diodes (for large fast transient) and pin diodes for levels continuous waves that exceed 100mW (+20dBm). I'm trying to preserve intermodulation performance, so the pin diodes when activated behave more like resistors, unlike the schottky, which will clamp the signal and degrade the overall intermodulation performance.

I think the circuit will work, but I was wondering if there is anything out there that is a bit more elegant.

thanks

Do you need a passive circuit ? If not you could try an RF detector driving a PIN diode attenuator. In front of it you can place a passive limiter that operates on very high levels and protect the system even if it is switched off.

To my knowledge PIN diodes only function above ~10...20 MHz, so your limiter can only utilize Schottky diodes.
You have not indicated what power levels are expected. In shortwave receivers you can use high-linearity MOSFET transistors and a fast AGC. For extreme overdrives, neon discharge tubes can "swallow" high power signals at receiver input.
Any non-linear element in receiver chain causes intermodulation. This cannot be avoided but by a better filtering.

In HF band up to 30MHz, the noise floor is so high that it is possible to use power transistor or FET front-end devices that are simply robust enough to take it.

Front-end overload from out of band signals are normally attenuated by some form of pre-selector filter or fixed bandpass filter. Again, the insertion loss of these is not normally important given the incoming noise level.

100mW is a small transmitter level. For a sensitive front-end to survive it, let alone use it as if it was a distant signal instead of one directly coupled into the near field of the antenna, there would need to be attenuation from a ALC-driven input attenuator. At these levels, it would be feasible to leave out the RF amplifier stage altogether, and feed it direct into a hard driven switched mixer ring. HF can now be directly sampled with various data converter, as with SDR (Software Defined Radio), though there are always great benefits in losing unwanteds with pre-selector filter.

As 100 nW ~2 V rms, just use two si diode in reverse parallel configuration (4 diodes), should give adequate linearity up 100 mV.
FWIW we used a minature toroid (~2mm diam) wound with 3 windings, 1:1 :5 (input, output, diodes) and used 5 si diodes in series on the diode winding. This way you could load 25 W of RF into the front end of the walkietalkie. Squadies used to test the sets by touching the aerials, one on RX, the other on TX ! As the power hit the core it got hot, reached its Curie point and then no coupling between the in and out windings.
Frank

No they can also work in the kHz range. For instance: http://www.avagotech.com/docs/AV02-1962EN. Of course P1dB and IP3 are very important parameters to take into account.

Thank you for the AVAGO pin diode infomation! So far I have not knwn that PIN diodes can work below 10 MHz.
Then yes, such PIN attenuator can be used for this SW appliction.