Frequency detuning with RF gain control

Hi, in my design, **broken link removed** I notice the next thing:
When SW1 is as shown (reflection ON), the RF gain at the emitter of the BJT does not affect the frequency that the receiver is tuned.
However, when reflection is swithed to OFF, the RF gain affects the frequency that the receiver is tuned.

I am truing to figure out why is that happening and how can if be corrected?

Adjusting the impedance of a choke by shunting it with a resistance is simple but unreliable. You are basically using the combination of a complex set of impedances in parallel with another complex set of impedances to control negative feedback in the BC549. It obviously works but the results may be unpredictable and unrepeatable.

However, the tuning change is probably due to the overall capacitance across the BC549 changing and having a similar effect as the varicap changing. Interaction between controls that adjust resonance, impedance and DC conditions in an oscillator circuit will inevitably change its frequency.

Brian.

betwixt said:

Adjusting the impedance of a choke by shunting it with a resistance is simple but unreliable. You are basically using the combination of a complex set of impedances in parallel with another complex set of impedances to control negative feedback in the BC549. It obviously works but the results may be unpredictable and unrepeatable.

However, the tuning change is probably due to the overall capacitance across the BC549 changing and having a similar effect as the varicap changing. Interaction between controls that adjust resonance, impedance and DC conditions in an oscillator circuit will inevitably change its frequency.

Brian.

Click to expand...

The weird thing is that, why this frequency pulling does NOT occur, if the reflexion is stitched to ON, no matter if I change the RF gain control then?
pulling only occurs when the reflexion is switched to off.
this is what I am trying to figure out.

Depending on what is connected to the "afout" point, there could be quite a lot of oscillator voltage from the FET fed to the base of the BC549 and it would change it's DC operating point. When SW1 is changed, that source of bias is removed. Bear in mind you are probably noticing a change of only 1pF or 2pF in tuning capacitance, it is very small.

Try adding a 1nF capacitor from the junction of the 10K fixed resistor and 100nF capacitor to ground. It will stop RF getting back into the BC549 but not stop the audio.

Brian.

betwixt said:

Try adding a 1nF capacitor from the junction of the 10K fixed resistor and 100nF capacitor to ground. It will stop RF getting back into the BC549 but not stop the audio.

Brian.

Click to expand...

I had done that at the very beginning (22nF) but I removed this shunt capacitor, as I thought the 3.3mH choke would be sufficient to block RF from the oscillator from feeding back to the front end amplifier.
I will add this capacitor again to see what happens.

However, the frequency is pulled when the reflexion is switched off i.e when no signal is fed back from the oscillator. I would expect that to be the other way (reflexion on). It seems that this unwanted bias is desirable.
Is that normal?

Ps. maybe I should better include an input RF attenuator (at the base) instead if changing the emitter resistor. This would achieve the same effect, max audio amplification and RF attenuation. Maybe the detuning due to emitter resistor variation would not exist then.

Ps. maybe I should better include an input RF attenuator (at the base) instead if changing the emitter resistor. This would achieve the same effect, max audio amplification and RF attenuation. Maybe the detuning due to emitter resistor variation would not exist then.

Click to expand...

Potentiometers tend to be unpredictable at high frequencies because of all their stray capacitance and track inductance but give it a try and see what happens. The other option that might work is to add a capacitor in series with the 'hot' end of the pot where it joins the emitter and choke. It will eliminate any DC changes due to voltage drop across the choke but use a fairly large capacitor (100nF for example) so there is little risk of making it resonate with the choke at frequencies you are interested in.

Brian.

betwixt said:

Potentiometers tend to be unpredictable at high frequencies because of all their stray capacitance and track inductance but give it a try and see what happens. The other option that might work is to add a capacitor in series with the 'hot' end of the pot where it joins the emitter and choke. It will eliminate any DC changes due to voltage drop across the choke but use a fairly large capacitor (100nF for example) so there is little risk of making it resonate with the choke at frequencies you are interested in.

Brian.

Click to expand...

Thanks!
What value of pot should I try for base attenuator? 1k, 10k, 100k?

What value of pot should I try for base attenuator? 1k, 10k, 100k?

Click to expand...

No more than 1K and preferably less but be sure to feed the antenna to the wiper and wire it on the antenna side of the 47pF capacitor or it will also short out the audio in reflex mode.

Brian.

betwixt said:

No more than 1K and preferably less but be sure to feed the antenna to the wiper and wire it on the antenna side of the 47pF capacitor or it will also short out the audio in reflex mode.

Brian.

Click to expand...

I thought so, so that the preamp sees a constant impedance at all potentiometer settings. So I fed the antenna to the wiper, one side of the carbon to the ground and the other to the capacitor. As predicted the frequency pulling was much much less!

I also placed a capacitor from the wiper to the antenna, is that really needed?

I also placed a capacitor from the wiper to the antenna, is that really needed?

Click to expand...

Not really, unless there is a possibility of DC reaching the antenna.
Wiring that way helps to keep a constant impedance at the transistor but don't forget the antenna itself presents an impedance which could be from a few Ohms to several K so near the top of the control it may still pull slightly.

Brian.