TheRadioBoard

I have built an FM Regenerative receiver with an HF pre-amplifier and a regenerative circuit. It is not a superreg, so I don't have any nasty noises and whistles and the reception is good.

Here is the schematic diagram of this receiver:"http://www.uploadarchief.net/files/download/fmontvanger.jpg

It is in facte the same design according to this schematic from B-Kainka:
http://www.uploadarchief.net/files/download/ukwaudion.jpg

Ik have improved this receiver.

I have connected the Q-multiplier (oscillator/ontdemper in the schematic), consisting of T2 and T3, onto the resonant circuit L1/C2 by means of a coupling coil of two windings. The result is that the receiver has become more selective and that it doesn't suffer from detuning of the oscillator circuit L2/C5 when I I adjust the tuning capacitor C2. The amount of gain of the RF-preamplifier (HF-voorversterker in the schematic) can be adjusted by the trimpotmeter R3. Unwanted oscillations which can occur when the gain of the RF-amp is too high, can be avoided on this way. I can also use a stopper resistor between the collector of T1 and the resonant circuit L2/C5 for avoiding unwanted oscillations.

By adjusting the potmeter P1, the amount of Q-multiplying of the resonant circuit L2/C5 can be adjusted. Whith P1 the reception of the desired station can be optimized. When the Q-multiplier is oscillating, the oscillator is synchronized to the signal coming from the resonant circuit L1/C2. The frequency variations of the FM signal are converted into amplitude variations by means of slope detection occurring in the circuit L2/C5.

The detector, consisting of J-FET T4, is biased in the non-linear region of its -Ugs/Id characteristic by the source resistor R7. Because R7 is so large, the -Ugs of the FET is near its cut off voltage. The signal is demodulated by the non-linear region of the characteristic. A detector like this one is called 'Drain bend detector'.

I'm building this receiver (improved, third schematic). I'm unsure how to construct the coils and RF transformer.

Should I use the measurements in the first schematic of the first posting? If so, what does CUL ϕ 0,5 mm, ϕ 3mm mean? Is that a 0.5 mm wire diameter and 3mm coil diameter?

Second, does your RF transformer in the third schematic have a core? If so, what kind and size of core? What is the size of the transformer coils and wire? Thanks in advance for the help.

Macrohenry

The primary side of L1 consists of 5 turns, the secondary side 2 turns. The windings have a space of about 2 mm between them. You put the secondary windings into the primary winding at the cold end. The coil L2 consists of 4 turns.

The coils don't have a core. But if you use a ferrite core, you should lower the amount of windings.



It means that the wire has a diameter of 0.5 mm and that the coil has a diameter of 3 mm.

Thanks. I built it and still have work to do. The output is very, very faint. I checked the oscillation with a portable FM radio next to it.

Part of the problem may be that I used transistors I had on hand, 2N2222 and MPF102.

In your diagram, is the collector of T2 connected to the base of T3? I did that and it's the only way I could get it to oscillate, and it oscillates only very close to the end of the pot. There seems to be very little middle ground, i.e. the oscillation just pops in.

I may need to bias the MPF102 differently to get it to detect better. More as I work on it.

Macrohenry

The collector of T2 is connected to the base of T3. It is very important to use RF transistors in the preamplifier and oscillator. The oscillator is a differential amplifier oscillator.

This circuit must be built upon a copper clad board with insulated pads, Manhattan style.

I'd built it dead-bug style and rebuilt it more to Manhattan style. It turns out the oscillation is in the RF amp. Also the Q multiplier is not working. I've not yet ordered the BF494 transistors, still using the 2N3904 which are supposed to be good up to 300 mHz. I don't know if they are the problem, but I do have a question about the Q multiplier.

It's arranged as a long tailed pair. However, when I look up Q multiplier circuits, the traditional one is a pair of FETs of opposite polarity. Yet the ones in the FM circuits above use the same transistors. How can this be accounted for?

Note that on your latest schematic, the labeling of T2 is obscured. It is also a BF494, right? And the base of T3 is not connected to the collector, right? Also, in the other circuits, the coil of the Q multiplier is in parallel with the capacitor to ground. In your improvement, it's in series, going to VCC. Why is it configured this way? Thanks in advance.

Macrohenry

@Macrohenry:


The base of T3 is connected to the collector of T2, both are BF494. See this schematic explaining this Q-multiplier setup. In this explanation are PNP transistors used. It is in Dutch.



Make sure that the input and output of the RF-preamplifier cannot see each other. They must be well seperated, otherwise it will oscillate. And with R3 you can adjust the gain of the RF-preamp. If the gain is too high, it will oscillate.

The capacitor C3 is parallel with L2. But because the rotor of C5 has to be connected to ground, the supply rail is connected to ground by C6. It is in fact in series with C5.

I'm having trouble visualizing this. In the Dutch diagram, C3 is in parallel with L1, as in the other circuits. In your version, L2 and C5 are not in parallel, but in series from VCC to GND. In your diagram, C6 and C5 are in parallel, not series. I'm starting to think that C5 should be connected to VCC instead of GND, putting it in parallel with L3. Is that correct?

Also, what is the function of 50pF capacitor C3? There are a lot of turns on its coil (L3), so what happens when you adjust C3? Thanks.

Macrohenry

C3 is for matching the aerial impedance to the input of the RF-preamp stage.

C6 and C13 are decoupling capacitors, they couple the Vcc to the ground for RF- and audio signals. The Vcc is then the same as the ground for the RF.

C4 has also the same function.

Thanks. So does this mean I can connect C5 to VCC instead of GND and it would work the same?

Macrohenry

I have built a different version of the receiver. It has a synchronous oscillater which is synchronised to the incoming signal.

http://www.uploadarchief.net/files/download/fm-synchroontuner%281%29.gif

At http://zpostbox.ru/vhf_radio_receiver.html there is a simple regenerative receiver with which you can receive FM broadcast signals. The regen stage is based upon a Hartley oscilator and works at only 1.5 V supply voltage.

Schematic VHF-regen:



At http://zpostbox.ru/simple_vhf_fm_radio.html and http://zpostbox.ru/fm_receivers_with_pll.html are FM receivers with synchronous oscillators which are locked onto the received signal from the antenna. The received signal is twice the oscillator frequency and is heterodyned with the second harmonic of the oscillator signal, the difference frequency between the the received signal from the antenna and the second harmonic of the oscillator is the LF signal.





Has anybody experience with this kind of receivers with regenerative stages which are synchronised to the incoming RF signal from the antenna?

I've built few of these synch detector RXs in the old days This is quite nostalgic piece of history for me .

These last two circuits are not regens, they are PLL DC receivers using the second harmonic of the LO for conversion. I believe their frequency pulling mechanism is different from a typical regen. In a regen the pulling (injection locking) is achieved via fast AGC action. In these receivers the frequency pull is the result of varying the transistor parasitic capacitance controlled by the detected beat frequency AF signal. When in lock the LO second harmonic is offset approximately 90 degrees from the carrier, the AF signal is proportional the phase offset variation from 90 degrees. The AF voltage controls the the LO frequency like in a PLL.

Note that in theory the AF signal should have a constant amplitude, proportional to frequency deviation (and transistor parasitics sensitivity to DC parameters) but not to the amplitude of the incoming signal. I do not recall how it worked on isolated weak stations, I was happy enough I could get 2-3 locals inside the concrete high rise building with a 2feet antenna.

BTW ГТ311 is a VHF germanium transistor and I remember it was much easier to get oscillating than a general purpose silicon KT315 in that circuit. The audio quality was quite good (as far as I can remember). When tuning between two strong nearby stations the tuning tend to snap from one station to another with some hysteresis. Tuning to a weak station near the strong one is nearly impossible with these due to the tendency to lock on the stronger carrier.

@Vladn,

What will happen when I take the oscillator frequency the same as the incoming RF frequency? Will it also have the same good result as with mixing the incoming RF signal with the second harmonic of the oscillator?