Transistor FM Superhet -Initial experiments

[transistor495]

G'day guys!

I've carried out some quick experiments after the arrival of my new RF Signal generator/counter. This is one of that. A transistorized FM superhet receiver which can demodulate and output broadcast band FM stations.

But the interesting thing is that I have started with a basic Medium Wave receiver circuit with 455KHz two stage IF and basic germanium diode AM detector. This one was a big success and I could receive the local AM stations clearly and apparently I got surprised a bit on how beautiful is the so called 'superheterodyne' principle!

The next stage was to remove the oscillator/mixer and detector part from the working receiver to actually convert it to an FM receiver. So I've added a simple charge pump FM demodulator consisting of 2 BC547's and a BF494 wired before the IF stages as a simplest mixer. Being connected with just a 5" antenna at the emitter of the transistor and Signal Generator output injected to the base in place of LO, let me say I'm quite surprised that the system could clearly receive couple of strong local FM stations nearby.

For 102.3MHz, I've injected around 102.9MHz to tune in the station clearly.

No tuned input circuits employed as of now and the whole circuit is powered using 3V(2xAA cells)

Next part would be to add an RF amp and LO so that to achieve a stand alone working FM receiver and I feel pretty confident that that would be so simple actually!

Below is the current status on breadboard, and you can see the signal generator outputting 102.9MHz used for ~455KHz down-converting for the receiver.

Cheers!

 

[Nigel Goodwin]

I'm amazed you've got anything out of it, breadboards aren't suitable for RF - and the layout looks really awful!

 

[transistor495]

I'm amazed you've got anything out of it, breadboards aren't suitable for RF - and the layout looks really awful!

Thanks for your warm concern Nigel

I would expect the breadboard may create problem while implementing the LO and RF stages because it can certainly affect the desired band coverage. Other than that, the current setup is totally running on the 455KHz IF stage with external oscillator input injected, it seems almost fine. When we move on to higher frequency stages things will become worse I agree. So I may sometimes try the LO part on a separate pcb.

Regarding the worst layout and look, hey this was a very quick experiment actually done for testing my new signal generator, though it worked

The audio output is a clean compromise comparing to my previous super-regen.

I'll try to publish an article with latest circuit once everything is done, and this one uses very cheap components and 455KHz IFT's can be salvaged from any AM tranny/IC radios.

This one seems very easy

 

[Nigel Goodwin]

I'll try to publish an article with latest circuit once everything is done, and this one uses very cheap components and 455KHz IFT's can be salvaged from any AM tranny/IC radios.

So you're taking a radio to pieces to build a radio? - don't you spot the obvious flaw in that?

Many years ago there was a small junk shop for a while in a small town called Darley Dale, and a friend of mine lived near by - so I popped in once. They had a bucket full of new IF transformers, I often wished I'd bought a pile of them - but I didn't

 

[audioguru]

Why do FM radios use wideband 10.7MHz IF transformers or crystal filters instead of the narrow band 455kHz ones used in AM radios??
Why do narrow band AM radios sound muffled with no high audio frequencies and are distorted, but half-decent FM radios sound extremely clear and produce almost all high audio frequencies without distortion??

 

[transistor495]

Why do FM radios use wideband 10.7MHz IF transformers or crystal filters instead of the narrow band 455kHz ones used in AM radios??
Why do narrow band AM radios sound muffled with no high audio frequencies and are distorted, but half-decent FM radios sound extremely clear and produce almost all high audio frequencies without distortion??

What you've told I agree with you AG, to be honest I haven't even think about this while building this one and was very quick. I felt I'm missing the treble frequencies a bit in the audio but it's distortion free and lower frequencies are good. This one should provide more than 10KHz of audio bandwidth. Later I'll move on to 10.7MHz IF.

Nigel, I've collected a bunch of new IFT's before and contains different types may be 10.7 also in them I need to check it out.

Another thing I'd be experimenting here is the Foster-Seeley discriminator

I'll add detailed comments later.

 

[audioguru]

In North America, AM radio stations are 10kHz apart so stations are not allowed to modulate with sound frequencies higher than 10khz to avoid beat frequencies in radios. Most AM radio stations use pre-emphasis equalization to boost audio frequencies from 2kHz to 5kHz to improve the intelligibility of speech through typical radios.

Wikipedia says that the audio response of the average AM radio is down 6 dB at 3 kHz and down
about 20 dB at 5 kHz. That is HORRIBLE audio frequency response like a telephone!

 

[transistor495]

Why do FM radios use wideband 10.7MHz IF transformers or crystal filters instead of the narrow band 455kHz ones used in AM radios??
Why do narrow band AM radios sound muffled with no high audio frequencies and are distorted, but half-decent FM radios sound extremely clear and produce almost all high audio frequencies without distortion??

FM use 10.7MHz because it provides better image rejection.

Dual conversion FM receivers use 10.7(or more) first IF and 455KHz second IF to achieve better selectivity. Still with hi-fi audio.

If anybody concerned about the speciality of '.7' in 10.7, the reason is that the image frequencies would never fall on any other station as it'll always end up with an even number and the station frequencies will always be an odd ending.

 

[audioguru]

Hi-Fi FM receivers do not use dual-conversion with 455 IF transformers. Low-Fi communications receivers with poor high frequency response and lots of distortion use dual-conversion.

 

[transistor495]

Being an audiophile like audioguru, I'd never make compromise on the audio quality of my receiver.

Currently I"m doing the front-end part of my 455KHz IF transformer(!) based FM wideband receiver

Let's see and modify later on to 10.7MHz.

 

[audioguru]

A 455kHz IF transformer is narrow-band for communications.
A 10.7MHz IF transformer is wideband for low distortion and wide frequency response for FM stereo receivers.

 

[Nigel Goodwin]

Dual conversion FM receivers use 10.7(or more) first IF and 455KHz second IF to achieve better selectivity. Still with hi-fi audio.

Sorry, complete nonsense - dual conversion is for communication receivers, the reason for it is to give low bandwidth voice quality only.

A 455KHz IF is FAR too low to give decent bandwidth or HiFi quality.

AG seems to have his own warped opinion about anything that isn't full HiFi quality, and while communication receivers can give excellent voice quality (which is their reason for being) HiFi it isn't, and doesn't want to be.

Coincidently, there's no 'magic' reason for 10.7MHz IF - it's simple a compromise giving reasonable selectivity, reasonable bandwidth, at a reasonable cost and ease of manufacture.

With modern filters (rather than pre-WWII transformers) you can make receivers that greatly exceed the performance of 10.7MHz IF receivers, in every aspect. Many more modern communication receivers get their greatly improved performances from far higher IF frequencies.

 

[audioguru]

I was never an amateur radio HAM.
In North America local telephone calls are free for any duration and sound better or the same as narrow band radio so I never used radio communications except for cell phones and FRS walkie-talkies.

 

[transistor495]

Well, not that easy

I was having a hard time getting this one working though, I had some results with severe overloading and lots of stability problems.

Within a 5 meter circle, I can move around to get a somewhat clear reception, I had noticed this with other portables also, but in my case its much severe, I've attached front-end RF amp though.

But all I can say now is I got the basic superheterodyne principle working for FM demodulation, but need more experiments to prove a nice working receiver. Also I'm running bit out of time.

So signing-off temporarily from this now and will update this thread after some time.

 

[Ramos]

G'day guys!

I've carried out some quick experiments after the arrival of my new RF Signal generator/counter. This is one of that. A transistorized FM superhet receiver which can demodulate and output broadcast band FM stations.

But the interesting thing is that I have started with a basic Medium Wave receiver circuit with 455KHz two stage IF and basic germanium diode AM detector. This one was a big success and I could receive the local AM stations clearly and apparently I got surprised a bit on how beautiful is the so called 'superheterodyne' principle!

The next stage was to remove the oscillator/mixer and detector part from the working receiver to actually convert it to an FM receiver. So I've added a simple charge pump FM demodulator consisting of 2 BC547's and a BF494 wired before the IF stages as a simplest mixer. Being connected with just a 5" antenna at the emitter of the transistor and Signal Generator output injected to the base in place of LO, let me say I'm quite surprised that the system could clearly receive couple of strong local FM stations nearby.

For 102.3MHz, I've injected around 102.9MHz to tune in the station clearly.

No tuned input circuits employed as of now and the whole circuit is powered using 3V(2xAA cells)

Next part would be to add an RF amp and LO so that to achieve a stand alone working FM receiver and I feel pretty confident that that would be so simple actually!

Below is the current status on breadboard, and you can see the signal generator outputting 102.9MHz used for ~455KHz down-converting for the receiver.

Cheers!

View attachment 60034

would you please post the schematic?


Here is what might be interesting for you too:

http://philsvalveradiosite.co.uk/philsvalveradiosite/pulsecountingfmreceiver_1.htm

http://www.vk2zay.net/article/250

http://www.youtube.com/watch?v=1q1NVZNNuiA

 

[transistor495]

would you please post the schematic?


Here is what might be interesting for you too:

http://philsvalveradiosite.co.uk/philsvalveradiosite/pulsecountingfmreceiver_1.htm

http://www.vk2zay.net/article/250

http://www.youtube.com/watch?v=1q1NVZNNuiA

Well, never noticed this reply ;-)

Those articles indeed looks interesting, but mine was to prove it with an easiest/simplest circuitry -perhaps I may need to conduct few more crude experiments and study. I'm ready to re-open the strategy -well..the curiosity never ends, sometimes the results will be horrible though

 

[audioguru]

A breadboard is usually not used for radio circuits because its stray capacitance between the many rows of contacts and between all the wires is far too much. The many long wires cause interference within the circuit and they pickup all kinds of external interference.
For many years, 10.7MHz crystal filters have been used in the IF of an FM radio instead of 10.7MHz transformers.

 

[transistor495]

I was getting into the track by doing a simple AM Superhet by using a single BF494 mixer+converter and a YS414 IF amp+detector. It works fine. This can be one of the simplest Superheterodyne designs. It has built in AGC provided by the chip. Audio is clean but it lacks the full punch which I assume because of YS414 has only output for a crystal earphone. I may need pre-amplification of the audio before sending to the power amp. But the current setup is still acceptable though.

Very quick experiment. Thanks for looking, cheers

View attachment 64770