[GuentherParticipant]

The common superhet(erodyne) receivers came up in the 1930ies (with radio tubes), and since November 1954 we have the transistorized superhets in pocket radios.

In this forum we had several topics with the introduction of AM receivers with this principle. Even the inventor of Lectron had plans in 1966 to develop blocks, see here.

Superhets with tubes for portable radios from 193x to 195x can also be built with Lectron, see here and here.

The most simpliest version is the 1-Transistor-Reflex-Superhet from ~ 1960.

Reflex-Superhets with 2 Transistors, even with a simple IC or for Shortwave  bands complete this overview.

But: how does it work? I will try to explain it with my simple words and a few samples:

The Lectron tuning block can be used as ONE L/C resonance circuit. But in regions with many radio stations – and in the evening, using a long wire antenna, or in the evenings in the colder seasons – it’s very difficult to “split” the sound from the stations to have a selective and sensitive reception of one station only.

One way out is to use a SECOND tuning block, like in this experiment from Raytheon:

 

Now you have to tune the first block, then the second, correct the first one again (fine tuning)  and so on… If you will change the frequency for another station the hole procedure starts again…

If the reception quality is not good enough, you may use MORE tuning blocks, like in this (upper) schematic:

 

 

It is rather impossible to tune 4 different L/C resonance circuits for an optimal result – but in the early radio years there were constructions like this (with radio tubes as amplifiers) – the idea of the superhet with a single-knob-tuning was born!

The principle:

–  ONE L/C resonance circuit for the radio frequency (RF) – with antenna or ferrite antenna for best reception.

–  a second mechanically coupled L/C resonance circuit for the oscillator frequency (OF), which is higher than the RF (difference = IF of e.g. 455 kHz/kc).

One transistor can work as mixer AND oscillator – the difference signal IF will be produced in IF 1.

Now may follow none (possible, but seldom) / one / two / three IF amplifier stages for the AM signal, but all stages on 455 kHz (!) This method don’t require any further tuning / adjustment – once a station is selected – all other “technique” is already implemented!

The last important stage is the Demodulator – the simpliest way is a Germanium diode that “rectifies” the IF frequency so that the HF part is suppressed and only the AF part passes and can be amplified: now you can hear the music, speech, whatever… clearly!

The above schematic is a reflex superhet – that means, that the AF part can be routed back to the base of the last IF amplifier stage and is amplified once more! Care must be taken how to lead the AF output to a following AF amplifier (otherwise it will find no exit…) Here it’s an AF input transformer, because a high-frequency signal will never pass such a high impedance “coil”! There are several tricky methods – it’s very important to know this ways…

If it’s not a reflex superhet there starts usually the AGC (automatic gain control) line. Depending on the different voltages after demodulation the base of the first IF amplifier transistor gets a different bias voltage – so a strong station will cause a lower amplification, a far-away station with low signal voltage will cause more amplification!

The following graphic shows again the three different frequencies and can make the general technique clearer:

Because of the different ratio the two lines (RF, OF) are unfortunately not absolutely parallel – this must be adjusted with the trimmer-C’s  parallel to the variable capacitors (and sometimes with additional so-called padder-C’s) to get the best possible synchronism.

But this is an one-time action, this procedure will be more explained in the Alignment Procedure for superhets in a new topic – together with my version(s) of a AM Modulator as a “helper tool” and a “private AM BC radio station” to listen to your favorite music with a DIY superhet and / or a Lectron receiver.

After knowing the required stages of a superhet and their jobs you are now able to unterstand any similar circuit.

This is a reflex superhet with only one IF stage.

This is a “normal” superhet with two IF stages and the AGC line.

The superhet components in detail:

The variable capacitor contains two of these with internal mechanically coupled rotor, and the trimmer-C’s. The pins are marked with (A) for antenna and (O) for oscillator. The middle contact is for GND. An AM BC pack has normally the values (rotors turned in) of about 160 pf (Ant) + 80 pF (Osc). If both packs have the same capacities, e.g. 2x 130 pF or 2x 345 pF this unit can be used for AM SW receivers, too – because we need for SW (nearly) identical  values. That may be clearer by this sample:

RF frequency (low) = 5.8 MHz/mc = start of the 49m band

OF frequency (low) = 6.255 MHz/mc (455 kHz/kc higher!)

So both frequencies are not so different as in the AM C superhet, and there this little difference will be”managed” by the oscillator coil! This coil has only a tiny difference to the RF coil / ferrite antenna. A ferrite antenna can be used up to about 12 MHz/mc (the ferrite material don’t allow a higher f).

The double pack may contain 2 another similar variable capacitors for FM use with about 2x 30 pF – don’t care…

The ferrite antenna receiving + coupling coil (here: samples for LW, BC and SW) must be movable for best possible alignment, therefore the Lectron antenna(s) of the legacy blocks cannot be used without change.

The second one from left is a suitable modification of mine and placed in the high block together with the double capacitor pack.

The oscillator and IF coils can be ordered – or taken from an older pocket radio:

Please note the normal colors: red = Osc, yellow = IF 1, white = IF 2, black/without = IF 3/Demodulator.

In commercial radios this colors may differ by about 5 % of all models. All these coils look like in this “explosion”:

Five pins for the connections (2 are the secondary coupling turns), and the IF coils must have the parallel capacitor(s) in the socket. The primary coil has a tap, important for transistor circuits – they need a lower impedance (in the collector path, < 5 kΩ). For circuits with tubes this tab will not be used, because tubes need high impendances (in the plate path, > 30 kΩ).

How to connect this coils?

In more than 90 % of all versions I guess the upper version will work right – the only difference is the oscillator coil pinout (mirrored?)! If the later circuit doesn’t work properly the coupling coil connections should be exchanged.

Further note the reduction of an oscillator capacitor of about 130 pF when you need only the half value: use a series capacitor Cs! The formula

Cs = (C1 x C2) / (C1 + C2)

is for the control of the required value, where C1 is the variable capacitor part (oscillator) and C2 the series capacitor. with 130 pF for C1 and C2 the result is 65 pF.

At the end I will insert a photo of a normal AM / FM transistorradio and it’s IF coils etc. to show how the different ferrit core colors should be interpreted. and to give some orientation. Unfortunately some people take a screwdriver and turn every part with a slot in the wrong opinion to get better reception… Please do it not in this way, too! Never turn on ferrit cores before you haven’t checked / verified the hole schematic for the “real” errors or defects!

The AM ferrit core standard colors we mentioned above, here are the “normal” FM colors: orange / green(ore without) / green / green (or without) / pink / light blue (ratio detector).

The alignment process is always backward (from right to left in the printed schematic) – with FM some experience and knowledge is required!

Discussions / questions welcome!

Best,

G.

• This topic was modified 4 years, 7 months ago by Guenther.
• This topic was modified 4 years, 7 months ago by Guenther.
• This topic was modified 4 years, 7 months ago by Guenther.

[MichaelKeymaster]

Hi Günther,

This is an amazingly detailed scholarly article about superhets!  So much that I did not previously know.

With the advent of digital radio, are there superhet like functions embedded in the ICs or has that technology been replaced?

Thank you,
Michael

[LectronFanModerator]

Thank you for this comprehensive information Guenther.

I have always wondered why a 455kHz IF was chosen, why this frequency ?

And I have also seen circuits using a different frequency.

Important to know when you would like to align a radio.

Also interesting to know that in a combination of FM and AM, the IF stages share the same IF transistor (this can also be seen in your last picture).

Ive been repairing a lot of these circuits at my former job and aligning those stages.

Sometimes the small capacitors inside the IF coils tended to drift.

This was primarily in TVs and I used an oscilloscope to find the resonant peak.

If these coils were out of tune, the TV station was out of tune and the auto search function didn’t perform well.

Many greetings.

[GuentherParticipant]

@ Michael:

there are about 17 different methods (digital radio, s. Wikipedia) worldwide in test – so I am tired since a couple of years to read or work with. Please wake me up when all technical problems are solved and the overall availability is garanteed! And how all the “old” (billions of) radios should be gathered and recycled! It’s an absolutely overkill to receive and prepare a radio station signal to listen to – when it takes hundred thousands (or millions) of transistor systems for processing. Why should I purchace and use a medium sized computing center and support programmer teams to code the logic, and wasting energy (voltage/current !) when it’s possible with one or two transistors like Lectron block #2461 ? We only have one planet to live on…

@ Frank:

The IF frequency was (and may be) on any frequency outside of the normal radio bands, e.g. 50…150 | LW | 300…500 kHz/kc. BUT: meanwhile (within the last 50…70 years) several frequency ranges are reserved for time signal (e.g. DCF 77.5 kHz/kc), weather, nautic (radio signals instead or additional to lighthouses at the coasts, think on the BRAUN T1000 station receiver with compass and two ferrit antennas for locating: 90 degrees crossed) on LW (long wave band). Strong LW stations like RTL may cause mixed radio signal products (second harmonic waves, together with special weather phenomena) so that since ~ 1954 the range 455…485 kHz was reserved for AM IF – the oscillator works like a transmitter, so it had to be on a frequency without near any radio station. It is possible to use ceramic filters (piezo resonators) in the IF amplifiers – all of these have an embossed or stamped number = IF frequency (on this photo the marking is sometimes on the unvisible topside):

American and Japanese radios all uses 455 kHz/kc – a far as I can remember.

The FM IF frequency started in Germany (and Europe?) with 6.75 MHz/mc, changed around 1960 to the – international – standard of 10.7 MHz/mc.

It’s no problem to use only one transistor for both IF frequencies, see this example of a pocket radio:

The collector is first connected with the FM IF coil, followed by the AM IF coil via tap. the FM coil inductivity is very low and is no remarkable influence for the AM coil.  The alignment should start with the FM part (all standard filters to highest noise between two stations, the two ratio filters at the end: symmetrical curve), then the AM part.

I had until now no deviations of the capacities – may be the warmth in TVs caused the drift – warmth always is a “good” way for premature aging…

G.

• This reply was modified 4 years, 6 months ago by Guenther.
• This reply was modified 4 years, 6 months ago by Guenther.

[LectronFanModerator]

Hello Günter,

This makes all sense now.
Still I wonder how some simple AM radios perform astonishingly well (2 IF amplifiers), while other using the same kind of circuit perform just OK.

I have noticed throughout my career that the European brands (ITT, Philips, Grundig, …) had very good reception while the Eastern Japanese or Chinese brands were not that good.

Has this something to do with the transistors used or could the IF and RF coils differ from quality ?

In the past years the AM band continues to become silent and the challenge is now to tune into anything ??

With the normal Lectron blocks, it seems an impossible job. With your constructed blocks there seems to be still a possibility.

Many greetings

[Author]

Posts