Here's my latest modified version of a home-brew IF transformer.

The primary idea here was to make the circuit a bit more easily mountable ( esp even under a small metal cover if desired) and to clean up the layout and simplify it.

I am using a 10-90 pf adj cap with approx 1.7MH coils.

I wanted an inductor that would work well with only the 10-90pf cap.

Because I didn't have a single 1.7MH coil I am using a 1-Mh in series with a 680uh coil.

By the way the adjustable caps cost 32 cents each and the coils were only 16 cents a piece.

Total cost was less than $1.50

http://www.futurlec.com/index.shtml
(Thailand)

Nice. What are you going to fabricate next?

:By the way the adjustable caps cost 32 cents each and the coils were only 16 cents a piece.
:
:Total cost was less than $1.50
:
:http://www.futurlec.com/index.shtml
: (Thailand)

No idea...lol

No need to top this Peter. Well done.

You have me thinking about 10.7 MHz IFs and descriminators. If I get time I will try it.

I am wondering if having a PC board done would be a good idea. Small runs are inexpensive now.

You hinted that input IFs and output IFs are different. The first was an input taking its input form a 12SA7. Did you resolve the problems with the output IF?

Best Regards,

Bill Grimm

Peter, I notice that you have a two coils in the primary and two in the secondary. Is that just so you can experiment with different values?
Doug

No Doug:

The two coils on each side(680uh+1000uh) add together to equal 1.7mh. It's just like resistors in series... I put the 2 inductors in series to get the total inductance I wanted.

If I could have found a single 1.7mh coil that is what I would have used.

Thanks Bill:
Here's a link to my thread in ARF where there's more info about the step down 2ndary winding.
http://antiqueradios.com/forums/viewtopic.php?t=86553

Awesome! I added the page to my favorites.

T.

The inductors / coils are physically close enough together that you get coupling between them. IF transformers depend on coupling / mutual inductance between coils. Too much coupling or mutual inductance equates to wide bandwidth and reduced selectability. Too loose coupling equates to loss of sensitivity and poor audio fidelity down the road. Off the shelf inductors tend to have low Q, which may explain why you do not get coupling from other stuff, and do not need to put the unit inside a metal shield. Inexpensive AM radios are not too picky, but IF transformer parameters are critical in the better receivers. One way to make a good IF transformer is to scrambe wind two coils of about 300 turns each on a BIC or similar pen tube, of # 33 to # 37 wire salvaged from a small transformer, and tune with 70 to 150 pf. You can experiment with the windings, and can find the center frequency out of circuit with an RF signal generator and scope, or sensitive meter, or a grid-dip meter. Feed signal in to top of coil through 10 to 100K resistor from a signal generator.

:Here's my latest modified version of a home-brew IF transformer.
:
:The primary idea here was to make the circuit a bit more easily mountable ( esp even under a small metal cover if desired) and to clean up the layout and simplify it.
:
:I am using a 10-90 pf adj cap with approx 1.7MH coils.
:
:I wanted an inductor that would work well with only the 10-90pf cap.
:
:Because I didn't have a single 1.7MH coil I am using a 1-Mh in series with a 680uh coil.
:
:

Thanks Ed I may try that too one day here soon.

The coupling here using these little coils however seems rather good and the simplicity kind of spoils me...lol

I tried a quick & dirty band pass test on this home-brew IF last night.

I ran the test first on a standard original IF and again on my home-brew IF and they are quite close.

What I did was to feed a steady fixed level of modulated 455khz into the grid of the 12sa7 Rf amp/mixer. I kept my frequency counter attached as well.

Then I measured the AVC voltage with my DMM. While watching the DMM I carefully varied the 455khz a bit until I saw the avc "peak".

Then I switched off the modulation and checked the freq counter to see EXACTLY what the input freq was... and it was right at 455khz...
OK.. so far so good.

Then I switched on the modulation again and listened to the audio tone as I again slowly varied the signal generator below 455khz until the audio on my radio speaker began to fade.

Then I switched off the modulation and read the freq meter which was just about 450khz.

Then I switched-on the modulation again and then varied the signal generator freq upward toward 455 and past it until I heard the audio just drop off again.
I switched off the modulation and read the freq counter as 460khz.

This indicated to me that the bandwidth is about 10khz for the original IF.

I performed the same test on my home-brew IF and came up with just a few khz wider on each side... indicating that the band width is probably about 12khz or so.

I don't know how accurate it is.. but it simply compares favorably to the standard IF ... so I feel I can't be very far off.

Peter, actually I think that 10-kHz bandwidth might be a bit too narrow. With the two sidebands, that would imply an audio response of just 5kHz, right?
Doug

:I tried a quick & dirty band pass test on this home-brew IF last night.
:
:I ran the test first on a standard original IF and again on my home-brew IF and they are quite close.
:
:What I did was to feed a steady fixed level of modulated 455khz into the grid of the 12sa7 Rf amp/mixer. I kept my frequency counter attached as well.
:
:Then I measured the AVC voltage with my DMM. While watching the DMM I carefully varied the 455khz a bit until I saw the avc "peak".
:
:Then I switched off the modulation and checked the freq counter to see EXACTLY what the input freq was... and it was right at 455khz...
:OK.. so far so good.
:
:Then I switched on the modulation again and listened to the audio tone as I again slowly varied the signal generator below 455khz until the audio on my radio speaker began to fade.
:
:Then I switched off the modulation and read the freq meter which was just about 450khz.
:
:Then I switched-on the modulation again and then varied the signal generator freq upward toward 455 and past it until I heard the audio just drop off again.
:I switched off the modulation and read the freq counter as 460khz.
:
:This indicated to me that the bandwidth is about 10khz for the original IF.
:
:I performed the same test on my home-brew IF and came up with just a few khz wider on each side... indicating that the band width is probably about 12khz or so.
:
:I don't know how accurate it is.. but it simply compares favorably to the standard IF ... so I feel I can't be very far off.

:Peter, actually I think that 10-kHz bandwidth might be a bit too narrow. With the two sidebands, that would imply an audio response of just 5kHz, right?
:Doug
:
Actually Doug I don't really know what it "should" be.
But I see that on the sweep-signal pictured here below for "Standard IF" the screen is calibrated to 10kc/per div and the peak looks like it isn't much wider than that.:
http://www.angelfire.com/electronic/funwithtubes/IF_Can-1.html

Doug:
Here's some more, very interesting and informative .. albeit somewhat-heavy reading on IF band-pass and such:
http://my.integritynet.com.au/purdic/if-amplifier-filters.htm

In the U.S. AM stations are spaced 10kHz apart, but because adjacent channels aren't assigned locally, I understand that broadcasters are permitted to audio modulate up to 10kHz - which requires a 20-kHz RF bandwidth to accommodate the two sidebands.

But your 10-kHz IF bandwidth (corresponding to 5-kHz audio bandwidth) should be fine for "talk" radio. And, a narrower IF bandwidth will improve selectivity.

I have an SStran AM transmitter which I can use to play LPs, FM, or CDs through an AM radio. The specs on the SStran claim its audio frequency response goes up to 20kHz, corresponding to a 40-kHz IF bandwidth! (Of course, the speakers in antique radios wouldn't be able to reproduce that high a frequency.)

I think you could gain music fidelity by tweaking the IF bandwidth up to, say, 15 kHz, without compromising selectivity much. You could do this by adjsting the coil spacing and reducing the Q.
Doug

:Doug:
:Here's some more, very interesting and informative .. albeit somewhat-heavy reading on IF band-pass and such:
:http://my.integritynet.com.au/purdic/if-amplifier-filters.htm

:Here's my latest modified version of a home-brew IF transformer.
:
:The primary idea here was to make the circuit a bit more easily mountable ( esp even under a small metal cover if desired) and to clean up the layout and simplify it.
instead you promote what is going to be listed on ebay. or do you sell these on ebay also?

:I am using a 10-90 pf adj cap with approx 1.7MH coils.
:
:I wanted an inductor that would work well with only the 10-90pf cap.
:
:Because I didn't have a single 1.7MH coil I am using a 1-Mh in series with a 680uh coil.
:

Who are you? Why are you AFRAID to use your REAL NAME???
...and to whom are you addressing your question?

If you are addressing ME ... I DID post that.. right? DUH.

Confused.