Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

455KHz bandpass filter requirement for superhet receiver

Status
Not open for further replies.

potcore

Electrical
Feb 26, 2007
28
0
0
GB
Hello

I have just designed and built the following simple AM superhet receiver.

3zc2uy8.jpg

and
359ff9i.jpg


It is just to receive the station on AM 1026KHz. The Mixer is a switching mixer and literally just switches the signal on/off at 216KHz using a 7555 timer and CMOS 4066 analog switches which are switched via buffer/inverter as required.
I am aiming to select the first IF on 810KHz, however my IF filter (a tuned common emitter stage) is obviously nowhere near selective enough. The receiver works but virtually certainly just detects the original station frequency of 1026KHz.

I hear there are specific steep-sided bandpass filters for 455KHz but am having difficulty sourcing these. –Would be very grateful if any reader knew of a source.
Also, since I will need to raise my mixer switching frequency to 571KHz, I wondered if readers may know if the CMOS 4066 analog switch’s that I’m using will switch at this rate? -the datasheet does not define this. –and whether a different switch would be more appropriate. Also, the “Practical Oscillator Circuits” book says that the 7555 is recommended “up to 500KHz and perhaps more” and wondered if anyone has ideas for a higher frequency square-wave oscillator (50% cycle)? –preferably one that has a slightly variable frequency so that I can ‘tune’ it to hit on the centre frequency of the IF filter.
 
Replies continue below

Recommended for you

One normally doesn't use those sorts of switches to do the mixing function. Diodes are common. Anything non-linear will probably get the job done.

Cheapest source for such filters might be inside an old $5 transistor radio. The little cans with the colourful slugs.

 
Thanks, i'll go along and get some of those filters. Reference the CMOS 4066 analog switches in the switching mixer -may i ask if you suspect that this mixer will not actually work as a mixer?
I would welcome any comments on this simple switching mixer, (no matter how derogatory!).
Some 'normal' mixers actually behave like switching mixers if the oscillator signal gets too big.
 
For a "simple superhet receiver" it looks rather over-complicated.

Those 10µF capacitors at the input and on the first two stages are a liability. You don't want gain at audio frequencies in the RF stages and the RF response of a 10µF is rubbish.

As I recall, there used to be a three pin chip from Ferranti (now Zetex) that did an AM radio. Why not use one of them?
 
Mixer theory often uses the word "switching". But that doesn't mean that you need to use a 'Switch'.

For high quality communications receivers, the designer will hopefully specify a high quality mixer (example: 'doubly-balanced') so that the receiver meets all kinds of detailed performance specifications. For a cheap AM radio, the designer might simply overdrive the next stage into non-linearity.

I've seens kids' walkie-talkies (a complete transmitter - receiver) all done with ONE transistor - ONE!!

 
Potcore:

450 KHz is too high for a potcore !!!!
VE1BL:
I have seen superheterodyne with two tubes
( 1 trode/heptode and one pentode. The triode is oscillator, heptode = mixer, pentode is IF amp
detector is a Ge diode then hexode nad pentode
-- in reflex -- the audio amp. )

The one transistor is superregenerative.




----------------------------
Please read FAQ240-1032
My WEB: <
 
Hi,

nbucska: Thankyou, i'll remember that...the varaible L in the above circuit is just a ferrite rod wrapped with ~180 turns of 0.28mm diam wire on a sliding former.

I'm still interested in any switching mixer comments (pse see cct diag above) which may be forthcoming. I have the thoery document for it but it seems a very different design than most others. I hope it works and doesn't just stuff the squarewave harmonics in with the signal.
 
Status
Not open for further replies.
Back
Top