AM Radio How to equalize upper and lower IF BW

[watchmaker]

I am using the Elenco AM/FM radio kit as a learning platform.  Leave aside it has the same endearing surprises as the ThaiKit CT.  I am over coming them (like replace the provided LM385 with a real one).

I have an IF BW of 462 and 439 kHz.  The instructions say the BW is acceptable.

However, I would like to get the response equal on both sides (this is a learning exercise after all).  I recorded the measured values of my resistors so that I can calculate expected voltages and currents for example.

What would need to be changed.  Or is it predetermined by the choice of IF xformers?  I do have a decadal capacitance substitution box.

I have found voltnut's youtube on this and he encountered many of the obstacles I encountered; and his ckt analysis helped me to extend my calculations.  But he accepts the BW and moves on. 

THANKS!

[wasedadoc]

I do not understand.

The IF frequency in an AM radio (530 to 1600 kHz or so) is typically in the region of 450 kHz. The IF bandwidth is about 10kHz.

For FM broadcast band (88 to 108 MHz) conventional design uses an IF of 10.7MHz and a few hundred kHz bandwidth.

What do you mean by "upper and lower IF BW"?

[BrokenYugo]

BW? Usually you just adjust the IF strip for max signal out of the detector or speaker with a modulated 455kHz test signal very lightly coupled in (you want the AGC at max gain), so every can is peaked to 455 in circuit. Bandwidth is baked in, I think in a single tuned transistor radio IF can this is set by the Q of the tuned circuit on the collector side, but that's about the limit of my understanding.

IIRC some old HiFi consoles had you optionally do a sweep alignment of the AM IF, so you could see and shape the passband by slightly detuning the cans or something to that effect. Something I've only read about, mostly an academic exercise with how little music is on AM anymore.

[CaptDon]

For best sensitivity you peak the cans for maximum recovered 455KHz. Bandwidth will be whatever it is. If a balanced sideband response is important (and it probably isn't) then you 'stagger tune' the I.F. with a sweep generator to give the desired I.F. passband. Usually requires an R.F. Detector probe to feed a D.C. analog of the I.F. output and a sweep generator that provides a sawtooth sweep voltage output. The sweep gen sawtooth output drives the horizontal input of the scope and the R.F. probe drives the vertical input. When all is set up correctly you get a trace on the scope that looks like the side view of an upside down bucket sitting on a floor.

[watchmaker]

5 kHz above and 5 below.

I think BrokenYugo and CaptDon gave me the answer.  After you align the 3 IF stages it is what it is.

Yes AGCi s peaked.

[watchmaker]

For best sensitivity you peak the cans for maximum recovered 455KHz. Bandwidth will be whatever it is. If a balanced sideband response is important (and it probably isn't) then you 'stagger tune' the I.F. with a sweep generator to give the desired I.F. passband. Usually requires an R.F. Detector probe to feed a D.C. analog of the I.F. output and a sweep generator that provides a sawtooth sweep voltage output. The sweep gen sawtooth output drives the horizontal input of the scope and the R.F. probe drives the vertical input. When all is set up correctly you get a trace on the scope that looks like the side view of an upside down bucket sitting on a floor.

BW? Usually you just adjust the IF strip for max signal out of the detector or speaker with a modulated 455kHz test signal very lightly coupled in (you want the AGC at max gain), so every can is peaked to 455 in circuit. Bandwidth is baked in, I think in a single tuned transistor radio IF can this is set by the Q of the tuned circuit on the collector side, but that's about the limit of my understanding.

IIRC some old HiFi consoles had you optionally do a sweep alignment of the AM IF, so you could see and shape the passband by slightly detuning the cans or something to that effect. Something I've only read about, mostly an academic exercise with how little music is on AM anymore.

Thank you.  This kit is definitely a relative of the ThaiKit Cure Tracer.  Lots of traps.

Yes, AGC is maxed.  And the 3 IF stages are peaked with good traces.

I am fine up until the oscillator.  The final BW is 4kHz which seems low.  Plus, I am having difficulty getting the 400Hz from the 540/1600kHz carrier modulated by the 400 Hz signal at 80%.  The frequency does change with the tuning capacitor, so I do not think that is the issue.

It does not help that the board is highly sensitive to 60 Hz even though I am in a basement, turn all LEDs off and attached the ckt ground to earth.  Using a battery to power the ckt, so not coming in from that.

May be my setup, but in addtiion to complaining about the noise pickup,  youtubers also seem to have difficulty at the point of final tuning..  Part of it may be the range of the trimmer on the tuner is too compressed which makes it hard to find the signals.

On the other hand, it has been a good learning platform.  Scope, sig gen, transistor biases, calculating ckt voltages and currents, etc.  So while I struggle, I learn.

[wasedadoc]

I am fine up until the oscillator.

There is very little before the oscillator, especially if it is a self-oscillating mixer with no RF amp in front of it.

[watchmaker]

I got myself sorted.  I failed to pay attention to my carrier frequencies.

I am in an AM dead zone, so I tried it on a trip to Keene NH today.  Picked up the college station loud and clear.  It is almost slient when there is no signal.

I may see if a radio freq amplifier will assist after the antenna.  Something to play with.

Going to put it aside for now so I can concentrate on my work (watch restoration) and my curriculum.  This was a nice break though.

Thanks to all!

Regards,

Dewey