I just put together an IF amp, an audio amp and an RF stage including a mixer. It's all built in stages for the sake of experimentation; the capability of varying the gain at several points along the way to the output has given rise to some questions. Here's the first question:

The IF is 455 kHz and an RF generator is used as the local oscillator. While playing with voltage levels, I've noticed the radio sounds best with the LO voltage is at a certain range of levels with respect to the strength of the rf signal. Is there a general rule of thumb with regard to the strength of rf to LO?

My mixer is an active one with a BJT in a common emitter configuration. The rf (an antenae) inputs to the base and the LO (tunable rf generator) inputs to the emitter. The collector connects to a transformer with a capacitor in parallel with the primary to form a single tuned tank to 455 kHz.

 

I just put together an IF amp, an audio amp and an RF stage including a mixer. It's all built in stages for the sake of experimentation; the capability of varying the gain at several points along the way to the output has given rise to some questions. Here's the first question:

The IF is 455 kHz and an RF generator is used as the local oscillator. While playing with voltage levels, I've noticed the radio sounds best with the LO voltage is at a certain range of levels with respect to the strength of the rf signal. Is there a general rule of thumb with regard to the strength of rf to LO?

My mixer is an active one with a BJT in a common emitter configuration. The rf (an antenae) inputs to the base and the LO (tunable rf generator) inputs to the emitter. The collector connects to a transformer with a capacitor in parallel with the primary to form a single tuned tank to 455 kHz.

The ARRL Handbook goes into great detail on all of this, and well worth perusing.

However, here are a few pointers on receiver design.

In general, for a PASSIVE mixer, you will have the best dynamic range when the ratio of L.O. power to incoming R.F. power is very high. For an active mixer, this is a bit different. You want the L.O. to drive the mixer ALMOST to saturation but not above. If you saturate the mixer with the L.O., you start to lose sensitivity.


Another factor is noise. Above a certain drive level, even passive mixers start creating noise, which can be counterproductive to dynamic range. This is usually not a factor until you start getting into the upper VHF or UHF frequency ranges. Below 30 MHZ, thermal agitation noise (the inherent noise of an antenna, by virtue of its mere existence) determines the noise floor.

Best practice for any receiver:

Use JUST enough r.f. gain before the mixer to overcome any mixer noise. MOST of the receiver gain should be in the I.F. stage.

The trend in modern receiver design is to get away from multiple conversions/ I.F. frequencies. In fact, the DC (direct conversion) receiver is considered the cleanest possible configuration, and is used in nearly all scientific and military H.F. applications. This has become possible only through the advent of very clean and stable synthesizers.


Eric

 

The ARRL Handbook goes into great detail on all of this, and well worth perusing.

However, here are a few pointers on receiver design.

In general, for a PASSIVE mixer, you will have the best dynamic range when the ratio of L.O. power to incoming R.F. power is very high. For an active mixer, this is a bit different. You want the L.O. to drive the mixer ALMOST to saturation but not above. If you saturate the mixer with the L.O., you start to lose sensitivity.


Another factor is noise. Above a certain drive level, even passive mixers start creating noise, which can be counterproductive to dynamic range. This is usually not a factor until you start getting into the upper VHF or UHF frequency ranges. Below 30 MHZ, thermal agitation noise (the inherent noise of an antenna, by virtue of its mere existence) determines the noise floor.

Best practice for any receiver:

Use JUST enough r.f. gain before the mixer to overcome any mixer noise. MOST of the receiver gain should be in the I.F. stage.

The trend in modern receiver design is to get away from multiple conversions/ I.F. frequencies. In fact, the DC (direct conversion) receiver is considered the cleanest possible configuration, and is used in nearly all scientific and military H.F. applications. This has become possible only through the advent of very clean and stable synthesizers.


Eric

Thanks, I read the ARRL article, but I'll review it. Maybe I'll understand it better this time.

You anticipated another question I have: How much rf gain should I use? Right now there is no rf amp -- just an active mixer as I described the post before this.

I plan to put am rf amp at the front end having a variable gain so that I can experiment with the rf and LO levels. I'll let you know what happens.

I'm curious as to how sensitive I can make the receiver. Right now it gets all but one of the local stations and some are weaker than others. I'd like to increase this as much as is reasonably possible.

It struck me that high signal gain in the later stages, such as the audio amp will amplify any noise right along with the signal. So I want as high a gain in the rf stage as possible without introducing noise.

Well, that's the plan, anyway. By the way my present antenna is foot long alligator lead.

 

Somewhere a while back I posted a piece on receiver gain, that is quite revealng. For a moderately strong shortwave signal, you need about 70 dB of overall gain to bring the signal up to reasonable audibility, with a decent set of headphones.

This 70 dB of gain can be distributed in any proportion between the R.F., IF, and audio stages.

For weak signal work, you need more on the order of 110 dB overall gain...in order to hear a signal down in the thermal agitation noise floor.


An "S9" signal at the antenna terminals has traditionally been calibrated to 50 microvolts. If the input impedance of the receiver is 50 ohms, this translates to a mere 50 PICOWATTS at the antenna! For a WEAK SIGNAL....i.e. an S1 signal, you're down in the FEMTOWATT region! It's amazing radio works at all, isn't it!

Eric

 

Somewhere a while back I posted a piece on receiver gain, that is quite revealng. For a moderately strong shortwave signal, you need about 70 dB of overall gain to bring the signal up to reasonable audibility, with a decent set of headphones.

This 70 dB of gain can be distributed in any proportion between the R.F., IF, and audio stages.

For weak signal work, you need more on the order of 110 dB overall gain...in order to hear a signal down in the thermal agitation noise floor.


An "S9" signal at the antenna terminals has traditionally been calibrated to 50 microvolts. If the input impedance of the receiver is 50 ohms, this translates to a mere 50 PICOWATTS at the antenna! For a WEAK SIGNAL....i.e. an S1 signal, you're down in the FEMTOWATT region! It's amazing radio works at all, isn't it!

Eric

I guess I accidently nailed it: My mixer front end has -3.7 dB, my IF stage (two amps) 53.5 dB and my audio amp 22 dB. All of which adds up to about 72 dB, but with the audio amp actually a variable.

As I said above I'm working on the actual front end: a CE amp fed by a transformer and with a transformer in the collector. The gain will be variable by using a variable resistor in conjunction with the bypass capacitor to ground at the emitter. This will enable me to make judgements about rf power vs LO power, vs audio power.

Also, I improved my antenna using a commercially made device and I get every station in town with ease. When all is working right, the sound is very good for AM, given its bandwidth.

Now comes the rf amp. I plan on making a broadband CE amp with transformers at the input and output for impedance matching. I also plan to put the antenna on a pole on the roof and feed it to my cable tv jack using coax. With this done I should be able to quantify the virtues a different gains here and there in the overall circuit. I'll let you know what happens.

As for radio being interesting, I agree with you whole heartedly!