3
\$\begingroup\$

I am currently designing an AM radio. I am utilizing a diode demodulator

enter image description here

(https://upload.wikimedia.org/wikipedia/commons/6/6c/Simple_envelope_detector.svg)

to extract the envelope of my carrier signal. I am using a 1N34A germanium diode, which has a forward voltage of 0.3 volts. I would like to know what are the constraints on the DC point at the input of the demodulator. I believe it is necessary to have an offset of 0 V so as to avoid following the input signal at the wrong moment, but I am not entirely sure. Could I benefit from a DC offset? I'm also wondering about the effects of loading on my demodulator, since I will be connecting a voltage amplifier stage before it. Thank you!

\$\endgroup\$
3
\$\begingroup\$

Often a small DC bias can help improve detection efficiency. Something around 100mV for a germanium diode. Often the DC bias is shared with the transistor bias such as in this design.

Simple Am receiver

A germanium diode actually starts conducting at much less than 300mV Germanium Diode characteristics so it can detect RF of only a few millivolts.

I'm also wondering about the effects of loading on my demodulator, since I will be connecting a voltage amplifier stage before it. Thank you!

Don't you mean after it, not before?

Yes you do have to be careful about loading the demodulator with the following stages. It should have a high input impedance relative to the RC filter after the diode.

What type of receiver are you designing? A TRF(Tuned Radio Frequency) or Superhet (Supersonic Heterodyne)?

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.