127kHz IF

[ukcol]

This question was asked in another tread.

Originally Posted by PJL

Anyone know the origins of 127KHz IF? Such a low frequency must have been recognised as a problem for image rejection.

Early superhets used different IF frequencies usually one in the range 110khz to 130khz.

A frequency chosen from this range was:

1] Low enough for a very high gain amplifier to be built without running into stability problems.

2] High enough to carry the modulation without distortion.

3] Not within the MW or LW range that might make it suffer from breakthrough from a strong radio station.


When manufacturers started selling sets with Short Wave the image rejection performance of a low IF was not good enough, hence the universal adoption of 465khz (or there abouts).

Some sets (like the Murphy A40C) kept the low IF and used double conversion for Short Wave, but that solution was too expensive for most sets.


Any good radio text will have the details and there is only relatively simple maths involved.

[PJL]

Hi Colin, thanks for the reply. I was thinking there would be a problem with MW but I can see your point that for SW it would be impossible.
LW 153-279kHz
MW 531-1620kHz
I guess another point is that a 127kHz IF does not need to be as selective as 465kHz to mask out adjacent stations. Perhaps the introduction of ferrite cores also played a part.

[ukcol]

Just as an interesting aside to this subject. Cossor produced a set (the 484) that used 2 IF frequencies.

On LW and MW it used 465khz but on SW it used 1363khz for the IF.

The set was interesting in other ways. It didn't have an IF amplifying stage which made it a "short" superhet but what was particularly unusual was that it had variable reaction applied to the IF transformer.

[Skywave]

Originally Posted by ukcol

When manufacturers started selling sets with Short Wave the image rejection performance of a low IF was not good enough, hence the universal adoption of 465khz (or there abouts).

Colin - the "110 kc/s IF feature" of early sets (& its consequent abandonment in favour of 465 kc/s) has always been one of those little things that I've always wondered about - as in why? Now I know - the fitment of SW bands on domestic sets.
Makes perfect sense to me - now - probably because it's obvious

I'm much obliged.

Al / Skywave.

[Skywave]

Originally Posted by PJL

I guess another point is that a 127kHz IF does not need to be as selective as 465kHz to mask out adjacent stations.

An IF of 127 kc/s will produce a more selective amplifier than an IF of 465 kc/s - and all other parameters being equal, significantly higher gain, too.

Al / Skywave

[ukcol]

Originally Posted by Skywave

An IF of 127 kc/s will produce a more selective amplifier than an IF of 465 kc/s..............

Al / Skywave

Hi Al

Yes, that fact is at least as important as those I mentioned.

['LIVEWIRE?']

Correct me if I'm wrong, but surely a high I.F. produces better SECOND CHANNEL rejection, whereas a low I.F. provides better ADJACENT CHANNEL rejection - e.g. Wanted station 1000kHz - I.F. 125kHz - Oscillator 1125kHz, thus a signal on 1250kHz will also produce an I.F. of 125kHz. If, however, the I.F. is(say) 1.6Mhz, the unwanted second channel freq. is now 4.2Mhz.
I always understood that this was why communications receivers used Double Conversion, sometimes with a 1st. I.F. of 10.7Mhz.

[Station X]

The difference between the wanted and image frequency is always twice the IF frequency or for a 1.6 MHz IF 3.2 MHz.

You are right in saying that a high IF gives better image frequency rejection and a low IF gives better selectivity and higher gain.

VHF transceivers are generally double conversion with a first IF of 10.7 MHz and a second IF of 455 KHz.

There are also triple conversion receivers with three different IF frequencies.

[johnn]

Hello all,
followed this thread with great interest as a Philips 727A has just arrived here.
It does have SW and uses 128kHz IF, where Bush use 465kHz & the same valves for their PB series sets.
Many thanks... John

[ukcol]

Hi John

Reference your Philips 727A.

Although this is a single conversion set with, as you say 128kHz IFs, I note that it uses a bandpass tuning circuit between the aerial and the FC.

This no doubt makes its image rejection performance, particularly on SW, much better than it otherwise would have been.

[Studio263]

I beleive the answer to the question is that a lower IF improves the "gain bandwidth product" of the IF stage as a whole.

The gain of all amplifiers that can remain in stability diminishes to unity (and below!) as frequency increases, obviously by lowering the frequency one moves away from that point.

To design an amplifier that gives stable gain at 470KHz is more difficult than doing it at 127KHz, one of the limiting factors is the stray capacitances in the IF transformers, valves etc. Lowering the IF effectively gives you more gain for free, that's a good thing.

Image rejection is of course a problem, that's why Philips included a tuned RF stage in most of their "low IF" models. As the RF stage has a certain amount of selectivity associated with it it helps to considerably reduce the appearence of unwanted images.

The last set I had with a low IF was a Philips 787AX (the "deluxe" monoknob, two parallel PEN A4's for the output, dynamic range compression, oversize loudspeaker with treble diffuser, TV sound band, magic eye, variable selectivity implemented by moving one of the IF cores, option of fitting a vibrator / rotary converter for DC use, hence the "X"...), the perfomance reflected the care taken in the design, truely excellent. I wish I'd kept it...

[johnn]

I will let you know how the 727A progresses. The wiring between its 3 separate chassis is rather too poor to use. An interesting one this one!
John

[Sideband]

The IF on my Philips 470A is 110Khz. Adjacent channel selectivity is excellent and puts many later sets to shame. The IF is very sharply tuned however so bandwidth suffers a bit.

I thought the low IF was chosen partly because the early bottles had trouble amplifying at higher frequencies !



Rich.