Audio detected in receiver from other frequencies

Hi, in my receiver, http://www.qrp.gr/allbander/index.htm I have noticed that sometimes, besides the useful tuned signals, I hear some stations that shouldn't exist in the passband. This happens only sometimes and it is independent from the tuning of the circuit. No matter if I tune it in different frequency, the same background station plays at very low volume though.

Initially I thought it would be intermodulation distortion by strong signals by the low dynamic range untuned RF preamplifier, so I replaced this stage with a high dynamic range untuned amplifier, dual gate mosfet running at 30mA. However, I could detect the same signals again, so I suppose it is not because of the dynamic range of the RF preamplifier.

Have you got any idea what could cause the problem? It is not annoying, as it is at very low level at the background when it happens, but I would like to know if I can fix this.

Non linearity in the base emiter junction detects any incomming RF that is strong enough. The resulting audio is then amplifed by the transistor, try dissabling the oscillator there should be no or very little difference in the unwanted signal level. What you have is an amplified crystal set.
Try a tuned input to reduce out of band signals, a wide band pre-amp however good will not help, the limitation is in the single transistor detector stage. It is a simple circuit you can not expect high performance.

Indeed, the sound and because it is unaffacted by tuning reminds me of an amplified crystal set, so I thought some strong RF input signal (when it occurs) is rectified by one of the two transistors.
So you think the problem caused on the preamplifier stage? This does not explain why in the other (untuned) high dynamic range preamp I have tested, the problem remains. Maybe the dynamic range was not enough though. The old Elektor article states 100dB, but it does not specify more info neither if it is IP2, IP3.

A preamplifier doesn't avoid intermodulation in the succeeding stage. The other option is that the interfering signal enters the circuit not through the antenna.

You would determine the frequency of the interferer and find out where it is demodulated in your circuit.

So the use of a higher dynamic range untuned preamp wouldn't help?
That's a good starting point. I will do so, when the interfering signal appears again.
If it is close to the frequency (for example strong commercial stations close to the HAM band), it would be very difficult to filter out effectively, or even at all.
It might be better to include an input RF attenuator, as many regenerative sets have, which however will attenuate the wanted signal as well...

Hm... now I am thinking whether the "rectification" is caused by the circuit or by the sound blaster (or the following audio stages I have tried). In other words whether RF that passes through the 100nF capacitor to the audio out is THEN rectified by the following circuits. I should place a 22nF to ground to ineffectively shunt RF, to see how it behaves, but as far as I remember, this was caused with the 22nF in place.

Hi, I know It has been posted a while, but I wonder, what if I implement the input RF preamplifier using a triode tube instead? Will the tube still "rectify" the incoming RF and act like an amplified crystal set?

The use of the transistor is not the problem and it will not be solved by switching to a triode. (Actually you may only add in a lot more problems by needing filament power as well as an HT supply for the triode - the transistor can work on lower voltages.)
The problem is that the unwanted signal is forcing its way into the rest of the circuit and you are then amplifying it. You must get rid of the out-of-band signals before you amplify.
Susan

It is quite possible that the interferring signal is not even coming thru the antenna. The next time you hear it, remove (or short) the antenna to see if it is still there. If it is, then you are just wasting your time with a better preamplifier. You will then need to look into better shielding of the different sections. If it goes away, you may need a band-pass filter in front of your preamp.

Sometimes a powerful signal from outside your band of interest can inject itself into your receiver at many different points - the trick is to identify where it is getting in and deal with it there. I agree that more preamplification, or even a tube, will probably not solve your problem - in fact, it could make it worse.

I once lived near an AM broadcast antenna that was religiously themed. One of my wired telephones would pick up this signal and blast it through to the ear-piece as clear as day (only the phone played it, it wasn't sent out to the other end). I called that phone my "jesus christ" phone, for obvious reasons. I had to tear into the phone and bypass that signal to the phone line ground.

I have tried that, when I disconnect the antenna nothing is heard. It is MW or SW broadcast stations because they are of foreign language. I do not see how the problem of the "amplifying crystal set" can be solved other than changing the topology of the circuit so that the RF preamplifier is no more used for audio preamplification as well. Maybe I could use the regenerative detector as an audio preamplifier, instead, as the input signals will be well filtered by the crystal prior to detection. Indeed when the reflex switch is swithed off, no unwanted signals appear.

Audio amplifiers ( especially wide-band ones ) sometimes can detect low frequency transmissions.Have you ever tried to shield the AF amplifier ?