Variable dead key question

[brandon7861]

I am not working on a radio, just looking for a little theory.

When I think of varying the dead key, I immediately think about varying the regulator that controls the collector voltage of the finals. However, i assume that if it were done that way, every time the collector voltage was adjusted, the modulation depth would also need to be adjusted (somehow asymmetrically).

I am hoping someone can explain to me the magic under the hood that allows the easy installation of a pot to vary the dead key without having to worry about transitioning from undermodulation to boxcars when using it. It doesn't seem like a simple thing to add properly.

Thanks!

 

[Eldorado828]

I am not working on a radio, just looking for a little theory.

When I think of varying the dead key, I immediately think about varying the regulator that controls the collector voltage of the finals. However, i assume that if it were done that way, every time the collector voltage was adjusted, the modulation depth would also need to be adjusted (somehow asymmetrically).

I am hoping someone can explain to me the magic under the hood that allows the easy installation of a pot to vary the dead key without having to worry about transitioning from undermodulation to boxcars when using it. It doesn't seem like a simple thing to add properly.

Thanks!

That's a good question and there a few true techs in here that can answer it.
I'm at full attention to learn about this one too.

 

[Shadetree Mechanic]

I think it would go from under modulation to box cars if the limiter was taken out?

 

[unit_399]

Brandon -
Go to my 858 site, and check mod #18. https://unit399.wixsite.com/858ssb. It allows the deadkey to be varied while still allowing full audio swing. Nomad developed this mod and it is applied to the 858SSb board. It can be adapted to most CBs with a few changes. The later UNIDEN SSb chassis (148GTL, 2000GTL) eliminate the modulation transformer, and direct-couple the audio to the driver/final allowing variable deadkey with full audio swing. Nomad's mod accomplishes the same thing.

- 399

 

[brandon7861]

Thanks unit_399! Looks like a lot of work went into that page! I have some studying to do!

 

[nomadradio]

I am not working on a radio, just looking for a little theory.

There is a big dividing line between AM-only and AM/SSB transmitters.

The sideband CBs have a circuit to "turn down" the carrier to the legal 4-Watt level. Whistle into the mike of a sideband radio. This is how much AM carrier it would try to deliver without a circuit to cut it back. Earliest 23-channel radios would just use a dropping resistor in line with the final's collector circuit. Most 40-channel radios had an adjustment, either a 15-Watt rheostat or a transistor circuit to drop the voltage.

Naturally a radio with the transistor carrier-set circuit is easiest to hijack as a carrier control. Some of them will 'swing' at any carrier level with only a simple mod. Other radios, like the 1978 TRC-457 RatShack model need extra persuasion.

AM-only transmitters are built to deliver that 4-Watt carrier and not much more. There is no need for the factory to include a control circuit, so if you want a carrier control, the entire control circuit and control both must be added.

Usual method is to reduce the collector-supply voltage to the driver only, using a capacitor to deliver full audio peaks with the steady DC voltage turned down.

Most common circuit uses a NPN Darlington power transistor between the modulated B+ and the driver transistor's collector choke. A Darlington transistor has a lot of gain, and reduces the current the control potentiometer must deliver to the transistor's base terminal.

Here's what we use.

73

 

[brandon7861]

Thank you to both nomad and unit 399! Very cool stuff!

 

[brandon7861]

Sorry guys, I guess I still don’t totally understand it. I redrew the schematic for the Uniden Washington AMC to see it better, but I am stuck on a couple things.

What I figure so far is that R99 and TR32 forms a voltage divider (when TR32 conducts) that attenuates the mic audio going into the mic amp. For TR32 to start conducting, TR31 must start turning on. For TR31 to start turning on, its base needs to be pulled below 7.3v. For that to happen, the emitter voltage at TR33 must drop below 0.3v (because its base is fixed at 1v). R124 and R125 form another voltage divider, this time sampling from the AF power regulator/driver collector supply and passing along 31% of it to the emitter of TR33. That being the case, I assume, and this is where I start to get confused, the power regulator needs to be providing 0.94v or less to R124 for the AMC to start activating. Going across two voltage drops (the darligton pair power regulator), that’s about 2.32v at the base of the regulator before the AMC starts waking up without the path of D42 and R122 influencing anything.

The part that I have trouble understanding is that if the AMC is already turning on before the audio amp swings low enough to turn off the power regulator (before D42 can conduct), what does the path of D42 do? I am going to guess that when the audio from the mic amp swings lower than 2.32v, that decoupling capacitor charges through the speaker and moves the DC bias point up just enough that the negative peaks don’t pinch off??? And so far I have only seen how negative peaks are controlled here, not positive ones. I am clearly missing something and I feel like there is also some RC time constant stuff going on that I don't understand. What am I missing?

Thanks!!!

 

[nomadradio]

This sequence begins with TR33. The base terminal is held at a constant 1 Volt, more or less. The emitter gets a divided-down copy of the modulated B+ that powers the RF driver and final courtesy of R125 and R125. TR33 remains turned off until the difference between the base and emitter terminals reaches about 0.6 Volts. The modulated audio waveform will drive the emitter of TR33 down to around 0.4 Volts when the negative side of the AM modulation gets near to 100 percent. 100 percent negative modulation is zero amplitude. For this half of the audio waveform, the lower voltage is the higher modulation percentage.

TR33 will only turn on during those brief downward peaks, and when it does it provides a current path to ground from the base of TR31. When this happens we get current from 8 Volts out the collector of TR31 and into the base of TR32. C90 smooths out the current pulses from those negative audio peaks. This smooths out the action of TR32, serving to reduce the mike audio level by shunting it to ground before it can reach the mike amplifier. TR32 is, in effect a variable resistor, controlled by its base current.

Whew!

TR33 is a comparator, triggered when modulation amplitude falls below its sensing threshold voltage. TR31 is a charge pump, feeding the attenuator transistor TR32.

This circuit has been the industry standard for sideband CB radios for decades. Uniden, Cybernet and RCI all use this circuit, albeit with minor changes.

73

 

[brandon7861]

Rewording my confusion...

When that variable dead key is turned down, the voltage at the regulator is obviously lower. What happens when the dead key is turned down so that the collector is lower than mid-rail? The audio coming in is now trying to drive the regulator more negative on negative peaks than it does positive on positive peaks because it is biased below mid rail. Simply attenuating negative peaks does nothing for the positive peaks that now need to work harder to push the regulator from, say, 3v to 12v, instead of from 6v to 12v.

So, if the audio limiter only controls negative peaks, I have to assume the D42-R122 path acts as a current source on negative peaks, both engaging the limiter and charging the coupling capacitor at the same time, on the first strong negative peak, and the effect of charging that bypass capacitor is to raise the bias until the audio no longer draws current through D42. Am I understanding the function of that D42 path correctly?

 

[nomadradio]

If the AMC is active when the carrier is turned down, you will totally shut off the mike audio if the carrier level falls below the threshold for the comparator transistor. If not, the audio level will be seen to fall off as the carrier is reduced. Various 10-meter radios have been made with two AMC trimpots, one for the high side of a carrier switch, the other one for the low side.

Safe to say a lot of operators with a carrier control on the radio also have the AMC bypassed or turned way down.

73

 

[guitar_199]

Brandon -
Go to my 858 site, and check mod #18. https://unit399.wixsite.com/858ssb. It allows the deadkey to be varied while still allowing full audio swing. Nomad developed this mod and it is applied to the 858SSb board. It can be adapted to most CBs with a few changes. The later UNIDEN SSb chassis (148GTL, 2000GTL) eliminate the modulation transformer, and direct-couple the audio to the driver/final allowing variable deadkey with full audio swing. Nomad's mod accomplishes the same thing.

- 399

399,

Just a simple message of "thanks"!!!!!

Somehow I have missed your 858 page on this journey.... and just "took a short tour"!!!!!

Thank you so very much for this "treasure trove" of info. So much in there!!!! I have a pair of TRC-458's and saw the section on R/S's "detuning" mods... and had NO IDEA that they did that.

But again, many thanks!!!!
Bob