Hello all,
I am in the process of installing the IF's that Warren worked on for me for the repair with removing the old silver mica caps and supplying new caps to solder on the outside pins. I appreciate your time and effort Warren.
Question is after the install i would like to do the alignment with a signal generator and "O" Scope. My scope is a TEK 100 MHz with many attributes so i am sure i can use it to do the alignment.
All that i have read indicates that i need a signal generator that sweeps that will produce the 455 KHz so i can see the proper signal on the scope. I have a Heath mod RF-1 generator that i do not think has this function. If i need a different signal generator with the sweep, what would be a good choice?
Thanks
Mitch

Mitch

Why would you need a signal generator that sweeps for AM radio alignment? Set the signal generator to 455 KHz and connect it to the mixer tube. Be sure your signal generator is calibrated as to frequency.

You can align a radio by monitiring AVC voltage but since you have a scope available might as well view the signal. View output of IF stage with your scope. Set IF Transformers for peak signal.

Norm

:Hello all,
:I am in the process of installing the IF's that Warren worked on for me for the repair with removing the old silver mica caps and supplying new caps to solder on the outside pins. I appreciate your time and effort Warren.
:Question is after the install i would like to do the alignment with a signal generator and "O" Scope. My scope is a TEK 100 MHz with many attributes so i am sure i can use it to do the alignment.
:All that i have read indicates that i need a signal generator that sweeps that will produce the 455 KHz so i can see the proper signal on the scope. I have a Heath mod RF-1 generator that i do not think has this function. If i need a different signal generator with the sweep, what would be a good choice?
:Thanks
:Mitch
:

:Mitch
:
: Why would you need a signal generator that sweeps for AM radio alignment? Set the signal generator to 455 KHz and connect it to the mixer tube. Be sure your signal generator is calibrated as to frequency.
:
: You can align a radio by monitiring AVC voltage but since you have a scope available might as well view the signal. View output of IF stage with your scope. Set IF Transformers for peak signal.
:
:Norm

Whoa! You should at least use a scope, sweep generator, spectrum analyzer, and proton beam detector. Then and only then can you be assured that the job is done corectly. Far too many underestimate the difficulty of alignment and the expertise required. That's why so many radios are seriously out of alignment. Even one ten thousandth Hertz IF offset can easily cause distortion on the level of the skin effect we all try so hard to eliminate in our speaker systems.

Please be more careful with the advice you give.
:
:

:Hello all,
:I am in the process of installing the IF's that Warren worked on for me for the repair with removing the old silver mica caps and supplying new caps to solder on the outside pins. I appreciate your time and effort Warren.
:Question is after the install i would like to do the alignment with a signal generator and "O" Scope. My scope is a TEK 100 MHz with many attributes so i am sure i can use it to do the alignment.
:All that i have read indicates that i need a signal generator that sweeps that will produce the 455 KHz so i can see the proper signal on the scope. I have a Heath mod RF-1 generator that i do not think has this function. If i need a different signal generator with the sweep, what would be a good choice?
:Thanks
:Mitch

A sweep generator will vary the output frequency back and forth across the IF, you sync the scope with the sweep and see a graphic plot of the exact bandwidth and the exact results of your adjustments, say from 440 to 470 kHz. I have used them on microwave and co-axial cable equalizaation at AT&T, but I believe that for an AM radio, you just tweak for maximum AVC Voltage (actually minimum, since it is negative). The bandwidths AT&T used were like 8 to 10 mHz wide, and it was impossible to equalize one without a sweep generator.
Lewis

::Hello all,
::I am in the process of installing the IF's that Warren worked on for me for the repair with removing the old silver mica caps and supplying new caps to solder on the outside pins. I appreciate your time and effort Warren.
::Question is after the install i would like to do the alignment with a signal generator and "O" Scope. My scope is a TEK 100 MHz with many attributes so i am sure i can use it to do the alignment.
::All that i have read indicates that i need a signal generator that sweeps that will produce the 455 KHz so i can see the proper signal on the scope. I have a Heath mod RF-1 generator that i do not think has this function. If i need a different signal generator with the sweep, what would be a good choice?
::Thanks
::Mitch
:
:A sweep generator will vary the output frequency back and forth across the IF, you sync the scope with the sweep and see a graphic plot of the exact bandwidth and the exact results of your adjustments, say from 440 to 470 kHz. I have used them on microwave and co-axial cable equalizaation at AT&T, but I believe that for an AM radio, you just tweak for maximum AVC Voltage (actually minimum, since it is negative). The bandwidths AT&T used were like 8 to 10 mHz wide, and it was impossible to equalize one without a sweep generator.
:Lewis
:
:
Hummm- All of the service info I've ever seen, and right now I'm looking at Marcus' "Elements of Radio Servicing", Corbin's "Antique Radio Repair", AND Audel's "Radioman's Guide", and ALL of them state that the AVC should be disabled, or the attenuator of the generator should be set as low as possible so as not to activate the AVC circuit. Inject the 455kc signal, with the volume control wide open, through a cap. Connect an ACVTVM with at least a 1 meg input across either the primary of the output transformer, or the speaker terminals. Aligning is an EXTREMELY delicate operation and if you're not experienced should only be attempted if ABSOLUTELY necessary. It is possible that the work Warren has done will not need alignment. If so, I would STRONGLY advise using some service info as a guide. It is VERY easy to do more harm than good if the alignment is done wrong.
Good luck-
Terry

:::Hello all,
:::I am in the process of installing the IF's that Warren worked on for me for the repair with removing the old silver mica caps and supplying new caps to solder on the outside pins. I appreciate your time and effort Warren.
:::Question is after the install i would like to do the alignment with a signal generator and "O" Scope. My scope is a TEK 100 MHz with many attributes so i am sure i can use it to do the alignment.
:::All that i have read indicates that i need a signal generator that sweeps that will produce the 455 KHz so i can see the proper signal on the scope. I have a Heath mod RF-1 generator that i do not think has this function. If i need a different signal generator with the sweep, what would be a good choice?
:::Thanks
:::Mitch
::
::A sweep generator will vary the output frequency back and forth across the IF, you sync the scope with the sweep and see a graphic plot of the exact bandwidth and the exact results of your adjustments, say from 440 to 470 kHz. I have used them on microwave and co-axial cable equalizaation at AT&T, but I believe that for an AM radio, you just tweak for maximum AVC Voltage (actually minimum, since it is negative). The bandwidths AT&T used were like 8 to 10 mHz wide, and it was impossible to equalize one without a sweep generator.
::Lewis
::
::
:Hummm- All of the service info I've ever seen, and right now I'm looking at Marcus' "Elements of Radio Servicing", Corbin's "Antique Radio Repair", AND Audel's "Radioman's Guide", and ALL of them state that the AVC should be disabled, or the attenuator of the generator should be set as low as possible so as not to activate the AVC circuit. Inject the 455kc signal, with the volume control wide open, through a cap. Connect an ACVTVM with at least a 1 meg input across either the primary of the output transformer, or the speaker terminals. Aligning is an EXTREMELY delicate operation and if you're not experienced should only be attempted if ABSOLUTELY necessary. It is possible that the work Warren has done will not need alignment. If so, I would STRONGLY advise using some service info as a guide. It is VERY easy to do more harm than good if the alignment is done wrong.
:Good luck-
:Terry

You are very correct, Terry, if you disable the AVC and tune for max output, or tune for max AVC, it is doing the same thing two different ways. You are also right in saying that one needs to watch somebody do a couple of alignments before attempting to do one by yourself.
Lewis
:

::::Hello all,
::::I am in the process of installing the IF's that Warren worked on for me for the repair with removing the old silver mica caps and supplying new caps to solder on the outside pins. I appreciate your time and effort Warren.
::::Question is after the install i would like to do the alignment with a signal generator and "O" Scope. My scope is a TEK 100 MHz with many attributes so i am sure i can use it to do the alignment.
::::All that i have read indicates that i need a signal generator that sweeps that will produce the 455 KHz so i can see the proper signal on the scope. I have a Heath mod RF-1 generator that i do not think has this function. If i need a different signal generator with the sweep, what would be a good choice?
::::Thanks
::::Mitch
:::
:::A sweep generator will vary the output frequency back and forth across the IF, you sync the scope with the sweep and see a graphic plot of the exact bandwidth and the exact results of your adjustments, say from 440 to 470 kHz. I have used them on microwave and co-axial cable equalizaation at AT&T, but I believe that for an AM radio, you just tweak for maximum AVC Voltage (actually minimum, since it is negative). The bandwidths AT&T used were like 8 to 10 mHz wide, and it was impossible to equalize one without a sweep generator.
:::Lewis
:::
:::
::Hummm- All of the service info I've ever seen, and right now I'm looking at Marcus' "Elements of Radio Servicing", Corbin's "Antique Radio Repair", AND Audel's "Radioman's Guide", and ALL of them state that the AVC should be disabled, or the attenuator of the generator should be set as low as possible so as not to activate the AVC circuit. Inject the 455kc signal, with the volume control wide open, through a cap. Connect an ACVTVM with at least a 1 meg input across either the primary of the output transformer, or the speaker terminals. Aligning is an EXTREMELY delicate operation and if you're not experienced should only be attempted if ABSOLUTELY necessary. It is possible that the work Warren has done will not need alignment. If so, I would STRONGLY advise using some service info as a guide. It is VERY easy to do more harm than good if the alignment is done wrong.
::Good luck-
::Terry
:
:
:You are very correct, Terry, if you disable the AVC and tune for max output, or tune for max AVC, it is doing the same thing two different ways. You are also right in saying that one needs to watch somebody do a couple of alignments before attempting to do one by yourself.
:Lewis
::
:
:
Lewis,
I'am on my own without the Forum, thats why I ask the questions and need the help as I go. Your support is very welcome to me.
Thanks
Mitch

:::Hello all,
:::I am in the process of installing the IF's that Warren worked on for me for the repair with removing the old silver mica caps and supplying new caps to solder on the outside pins. I appreciate your time and effort Warren.
:::Question is after the install i would like to do the alignment with a signal generator and "O" Scope. My scope is a TEK 100 MHz with many attributes so i am sure i can use it to do the alignment.
:::All that i have read indicates that i need a signal generator that sweeps that will produce the 455 KHz so i can see the proper signal on the scope. I have a Heath mod RF-1 generator that i do not think has this function. If i need a different signal generator with the sweep, what would be a good choice?
:::Thanks
:::Mitch
::
::A sweep generator will vary the output frequency back and forth across the IF, you sync the scope with the sweep and see a graphic plot of the exact bandwidth and the exact results of your adjustments, say from 440 to 470 kHz. I have used them on microwave and co-axial cable equalizaation at AT&T, but I believe that for an AM radio, you just tweak for maximum AVC Voltage (actually minimum, since it is negative). The bandwidths AT&T used were like 8 to 10 mHz wide, and it was impossible to equalize one without a sweep generator.
::Lewis
::
::
:Hummm- All of the service info I've ever seen, and right now I'm looking at Marcus' "Elements of Radio Servicing", Corbin's "Antique Radio Repair", AND Audel's "Radioman's Guide", and ALL of them state that the AVC should be disabled, or the attenuator of the generator should be set as low as possible so as not to activate the AVC circuit. Inject the 455kc signal, with the volume control wide open, through a cap. Connect an ACVTVM with at least a 1 meg input across either the primary of the output transformer, or the speaker terminals. Aligning is an EXTREMELY delicate operation and if you're not experienced should only be attempted if ABSOLUTELY necessary. It is possible that the work Warren has done will not need alignment. If so, I would STRONGLY advise using some service info as a guide. It is VERY easy to do more harm than good if the alignment is done wrong.
:Good luck-
:Terry
:
Terry,
I am 82 with a new gift of life, I fear nothing about these radios except that I am ignorant about them but want to bring them back to their beautiful sound. Your support enabled me to restore my first radio, you did not quit on my ignorance. We got the radio working and I listen to it often. No 100 foot wave antenna.
Watch my posts and keep helping.
Bless all of you on this Forum, you are one of my reasons to keep posting.
Mitch

Hi Mitch. Your Heathkit signal generator is just fine for the 455KC alignment. The only thing is how well the signal generator is calibrated. The dial pointer on it may not be right on. One way to check the 455KC calibration, take a known good AM transistor radio, tune to a blank spot (ether end of dial) make a loop of wire connected to the signal generator output. Set dial of generator at 455KC RF Modulated output. Listen to the AM radio. Adjust dial pointer on signal generator till maximum tone is herd on the transistor radio. Mark that spot on the generator dial, as being the true 455KC. (assuming the transistor radio is a 455KC IF)

Mitch,
A sweep gen is an absolute must for alignment of an FM IF subsystem, where critical adjustment actually shapes the pass band (-3db @+/-75kHz) response of the IF transformers for the best audio/stereo reception. Unless you plan to stagger tune an AM IF subsystem, the sweeper is a waste, just to peak the transformers @455kHz. Assuming you have an accurate source of modulated 455kHz. Most scopes have a direct input impedance of 1 meg, a few picofarads parallel capacitance, and with an appropriate probe 10 meg, and a few picofarads. Ideally, you can monitor AVC on the scope (DC mode) without loading the AVC circuits, while tweaking the coils, but keep in mind that there will be a small delay in AVC response, determined by RC time constant of the AVC components. Don't use a DMM or VOM for this measurement. I would think that delay to be almost imperceptable by the human eye. Using AVC eliminates attaching meter or scope to RF circuitry, and in some cases loading/altering the results.

marv

:Hello all,
:I am in the process of installing the IF's that Warren worked on for me for the repair with removing the old silver mica caps and supplying new caps to solder on the outside pins. I appreciate your time and effort Warren.
:Question is after the install i would like to do the alignment with a signal generator and "O" Scope. My scope is a TEK 100 MHz with many attributes so i am sure i can use it to do the alignment.
:All that i have read indicates that i need a signal generator that sweeps that will produce the 455 KHz so i can see the proper signal on the scope. I have a Heath mod RF-1 generator that i do not think has this function. If i need a different signal generator with the sweep, what would be a good choice?
:Thanks
:Mitch
:

:Mitch,
:A sweep gen is an absolute must for alignment of an FM IF subsystem, where critical adjustment actually shapes the pass band (-3db @+/-75kHz) response of the IF transformers for the best audio/stereo reception. Unless you plan to stagger tune an AM IF subsystem, the sweeper is a waste, just to peak the transformers @455kHz. Assuming you have an accurate source of modulated 455kHz. Most scopes have a direct input impedance of 1 meg, a few picofarads parallel capacitance, and with an appropriate probe 10 meg, and a few picofarads. Ideally, you can monitor AVC on the scope (DC mode) without loading the AVC circuits, while tweaking the coils, but keep in mind that there will be a small delay in AVC response, determined by RC time constant of the AVC components. Don't use a DMM or VOM for this measurement. I would think that delay to be almost imperceptable by the human eye. Using AVC eliminates attaching meter or scope to RF circuitry, and in some cases loading/altering the results.
:
:marv
:
::Hello all,
::I am in the process of installing the IF's that Warren worked on for me for the repair with removing the old silver mica caps and supplying new caps to solder on the outside pins. I appreciate your time and effort Warren.
::Question is after the install i would like to do the alignment with a signal generator and "O" Scope. My scope is a TEK 100 MHz with many attributes so i am sure i can use it to do the alignment.
::All that i have read indicates that i need a signal generator that sweeps that will produce the 455 KHz so i can see the proper signal on the scope. I have a Heath mod RF-1 generator that i do not think has this function. If i need a different signal generator with the sweep, what would be a good choice?
::Thanks
::Mitch

Oh, yes, I forgot to say keep the AVC Voltage as low as possible by using a low input signal, or you will be fighting yourself if you let the AVC start controlling the gain of the radio.
Lewis
::
:
:

Here's my 2 cents.
Well yes, it pays to read the alignment instructions and to think about what you are attempting to do.. YES.. of course always.

But on the other hand it isn't that God-Awful scary either ..and in practice it can actually be quite simple.

Remember that the goal here is to trim the IF filter transformers for the optimum operation at the designed IF freq ( typically 455khz) ..right?

1.) So the easiest way is to simply inject a 455khz signal ( modulated with say 400hz.. so that you can hear it)
into the 1st IF mixer stage grid and then listen for the tone while tweaking the IF can trimmers to achieve a max or peak tone.

2.) Another method is to inject the 455khz modulated signal, then max the volume control and place an analog meter on the Output (speaker )transformer 2ndary and tweak for max... but that will be very loud and drive you nutZ!!

3.) If I want to use my scope and my frequency counter, then I first set my signal generator to 455khz (un-modulated) so as not to confuse the frequency counter, and slowly turn the dial till my freq counter reads exactly 455khz.

That gives me a good feeling that I'm actually using exactly 455khz.

Then I set the sig generator to MODULATED and inject the signal into the 1st IF mixer stage grid.

Again I listen to the signal peak as I tweak the IF trimmers.
.... BUT if I really want to be sure I've tweaked it as perfectly as I could have.. then I'll bring my scope into play.

I hook the scope to the AVC line and set it to DC.
This will give a trace line on the scope showing below the zero-volt center-line at about minus 4 to 8 volts depending on the radio.

I do not need modulation as I will NOW not be listening with my ears anymore ....but seeking a dip on the scope trace line.

So I turn off the signal generator's modulator and only inject the (un-modulated) 455khz RF IF signal into the grid of the 1st IF mixer stage.

At this time the scope should be showing a trace down at about -4 or -6 volts.

Then I simply tweak the IF trimmer caps to see the AVC signal voltage dip toward the lowest point and leave it there.

Doing this for both IF cans starting with the 2nd IF first .....then the 1st IF and watching the AVC dip will insure that the IFs are properly tuned for 455khz.

:Here's my 2 cents.
:Well yes, it pays to read the alignment instructions and to think about what you are attempting to do.. YES.. of course always.
:
:But on the other hand it isn't that God-Awful scary either ..and in practice it can actually be quite simple.
:
:Remember that the goal here is to trim the IF filter transformers for the optimum operation at the designed IF freq ( typically 455khz) ..right?
:
:1.) So the easiest way is to simply inject a 455khz signal ( modulated with say 400hz.. so that you can hear it)
:into the 1st IF mixer stage grid and then listen for the tone while tweaking the IF can trimmers to achieve a max or peak tone.
:
:2.) Another method is to inject the 455khz modulated signal, then max the volume control and place an analog meter on the Output (speaker )transformer 2ndary and tweak for max... but that will be very loud and drive you nutZ!!
:
:3.) If I want to use my scope and my frequency counter, then I first set my signal generator to 455khz (un-modulated) so as not to confuse the frequency counter, and slowly turn the dial till my freq counter reads exactly 455khz.
:
:That gives me a good feeling that I'm actually using exactly 455khz.
:
:Then I set the sig generator to MODULATED and inject the signal into the 1st IF mixer stage grid.
:
:Again I listen to the signal peak as I tweak the IF trimmers.
:.... BUT if I really want to be sure I've tweaked it as perfectly as I could have.. then I'll bring my scope into play.
:
:I hook the scope to the AVC line and set it to DC.
:This will give a trace line on the scope showing below the zero-volt center-line at about minus 4 to 8 volts depending on the radio.
:
:I do not need modulation as I will NOW not be listening with my ears anymore ....but seeking a dip on the scope trace line.
:
:So I turn off the signal generator's modulator and only inject the (un-modulated) 455khz RF IF signal into the grid of the 1st IF mixer stage.
:
:At this time the scope should be showing a trace down at about -4 or -6 volts.
:
:Then I simply tweak the IF trimmer caps to see the AVC signal voltage dip toward the lowest point and leave it there.
:
:Doing this for both IF cans starting with the 2nd IF first .....then the 1st IF and watching the AVC dip will insure that the IFs are properly tuned for 455khz.
:
:
I am printing all of these posts, seems that i can do what i need to do with what i have. I want this radio to be aligned as well as possible for all the work Warren and I have put into it. Plus this is a hugh learning curve for me. If I make mistakes I will call upon you for help. I am almost blind so this repair will take some time for me. With big Mags I can do the install as long as I don't break one of the small wires in the can's by mistake when making connections.
As far as I can tell my signal gen is about 10 Hz off cal. May ask for help on this as there does not seem to be any concrete way to calibrate and be absolutely sure of the final cal.
Appriciate all of you,
Mitch

Go for it. All you need is a regular signal generator, and either your ears or an ac voltmeter. Don't believe any nonsense about how complicated or critical am alignment is. Your radio isn't going to explode if you make a mistake.

If you aren't satisfied when you get finished, you go over it again. Just take your time and you'll see how easy it really is.

A 10 Hz error is nothing to be concerned about. Even if your generator is off by a couple of Khz, the only thing you might notice is some tracking error. You can do a fair job of checking your generator by zero beating a radio station. For instance, if you want to check the 455 signal, you would use the second harmonic of 910 and beat against a station at 910. If there is no station at 910 in your area, use one that is close to two (or three) times a frequency that is close to the frequency you want to check.

One other thing. If this is an ac/dc set you should be using an isolation transformer to power the set.

Have fun. Remember that this is a hobby.

One other, other thing. Do research on your own. There is wealth of real information on the web. This is a hobby site and there is often a lot of good data here. But always keep in mind that it is a hobby site.

::
:I am printing all of these posts, seems that i can do what i need to do with what i have. I want this radio to be aligned as well as possible for all the work Warren and I have put into it. Plus this is a hugh learning curve for me. If I make mistakes I will call upon you for help. I am almost blind so this repair will take some time for me. With big Mags I can do the install as long as I don't break one of the small wires in the can's by mistake when making connections.
:As far as I can tell my signal gen is about 10 Hz off cal. May ask for help on this as there does not seem to be any concrete way to calibrate and be absolutely sure of the final cal.
:Appriciate all of you,
:Mitch
:

Don't see your logic about using a scope to monitor the AVC? Wound it not be better to just look at the real IF wave form? As you say, the AVC voltage is -4 to -6 volts. That sure means the AVC is cutting in. You would have to keep lowering the RF output of the generator as it comes into alignment. thus chasing the AVC voltage up and down to find its lowest point with the least signal input. With a VTVM on AC across the speaker, the volume control really does not have to set maximum. The RF modulated output just cracked enough to hear the tone. Adjust for the highest reading on the meter.

Hi Warren:
Yes the AVC is active and yes the IF signal should be as low as possible.
.. but since you are peaking the IF for optimal BAND-pass the AVC will become more negative as the IF transformer is optimized.

So it is really easy to watch the scope trace dip at that point.

If you want, just to give yourself some additional confidence, you can watch the scope trace on the AVC line while your ears listen for an audio peak as well...if you'd like.

You'll see that the scope on the AVC line is a little more accurate then the ears.. but that's up to you.

With the AVC line method...the IF modulation isn't needed either, so that can often make the IF signal more accurate and easier to tweak.

Peter,
Whoa, wait a minute here. Doesn't the AVC circuit strip the RF (455kHz) from the signal, using only the resultant audio to generate a slowly varying AVC control voltage, or is it an RMS value of the incoming 455kHz wave?? If the former, then modulation is req'd.

marv

:Hi Warren:
:Yes the AVC is active and yes the IF signal should be as low as possible.
:.. but since you are peaking the IF for optimal BAND-pass the AVC will become more negative as the IF transformer is optimized.
:
:So it is really easy to watch the scope trace dip at that point.
:
:If you want, just to give yourself some additional confidence, you can watch the scope trace on the AVC line while your ears listen for an audio peak as well...if you'd like.
:
:You'll see that the scope on the AVC line is a little more accurate then the ears.. but that's up to you.
:
:With the AVC line method...the IF modulation isn't needed either, so that can often make the IF signal more accurate and easier to tweak.
:
:

Nope!
(Guess it's time to go beck and check up on theory agin...lol)
I've done it many times with NO modulation. Try it and see for yourself.

The IF doesn't need to be modulated to read voltage on the AVC line. And pure audio happens at or after the 1st audio stage.

The RF is stripped off later at the RF bypass cap usually at the output of the 1st audio where it feeds the grid of the (50L6) audio amp tube.

Here's a little primer as a reminder:

How AVC Works

"AVC circuitry is designed to keep the carrier-frequency potential at the 2nd detector diode stable by dynamically controlling the gain of the RF and IF stages as signal strength increases or falls. This is done through a d-c network beginning with the AVC diode and/or 2nd Detector circuits and feeding back to the control grids of the RF and IF amplifier stages and, on occasion, to the converter tube. In the absence of any signal, AVC circuitry typically develops between zero and -2 volts and has no effect on the IF or RF gain. As stronger signals are encountered, the AVC responds by developing a bias that becomes increasingly negative. This increased bias reduces the amount of current passed by the RF and/or IF tubes thereby reducing the gain in each stage. The typical maximum AVC bias developed in most sets is either -8 or -25 volts, although there are many variations."

"The AVC diode is typically found in the same tube as the 2nd Detector diode. In some sets the two functions are housed in a tube that is strictly a duo-diode type, almost always a 6H6 or 6H6G. Most sets used tubes that incorporated this duo-diode into the same envelope as the 1st Audio Amplifier triode, using tubes such as the 75, 6Q7G, 6SQ7, or 12SQ7. There are many variations of detector circuit architecture including triode detectors found in the earliest superheterodyne sets. The majority of sets you’ll encounter however will have a circuit similar to the one shown above.

How the AVC voltage is developed is relatively simple. The AVC voltage itself is developed across the AVC diode load resistance. This is the cumulative resistance of both R-26 and the series-connected volume control. In the absence of a radio signal, little or no current passes through the AVC diode. Little or no AVC diode current means little or no current through the AVC diode load resistance. From Ohm’s law, little or no current through a resistance produces little or no voltage drop. Hence, when the AVC diode isn’t conducting, the AVC voltage is at a minimum.

As signal strength increases, the AVC diode conducts correspondingly. This causes a current to pass through the AVC diode load resistance. From Ohm’s law, a voltage is then developed across the load resistance and will become increasingly negative. This negative AVC voltage is fed back to the grids of the tubes in the IF and/or RF stages thus reducing how much these tubes conduct. By reducing how much these tubes conduct, the signal level delivered to the AVC diode is stabilized. This is the key regulating function for AVC.

In addition to the AVC diode load resistance, many sets utilize more than one resistor in series with the AVC circuit to provide isolation between stages. This is part of the function of resistor R-28. Another fundamental aspect of AVC is the time delay for AVC to change in response to signal change. The circuit elements for this important part of AVC are not shown in the AVC circuit diagram."
REF: http://home.pacbell.net/philbert/tuning_eye/howavc.htm