Frequency Counter / Monitoring AM Carrier Only

[K4RT]

I would like to monitor the AM carrier of my Heathkit TX-1 with a frequency counter, perhaps at the RF drive stage, rather than using the counter's antenna to pick up the RF when
transmitting, which works fine, but also picks up the varying sideband frequencies when the carrier is modulated on air.  I would like to be able to keep an eye on the carrier frequency only when transmitting, and touch it up occasionally with the front panel VFO knob.  I can do that now if I pause while speaking so there's no modulation, but I would like to have the counter just reading the carrier at all times.

Have any of you guys or gals done this?  Where and how did you tap into or couple to the RF stage?

It seems practical to me to tap into or couple to the RF stage, perhaps running a shielded cable to a rear panel BNC connector, but I have not found any information explaining how to do it in my radio books or on the web.

I'm not interested in a digital VFO for the TX-1.

[AB2EZ]

There are several approaches to this... all of which have some shortcomings.

You should verify that modulated RF from the output stage is not feeding back into the VFO. If it is, there will be phase modulation accompanying the AM... and, depending upon the details, this can cause the counter to jump around. This effect is greatest when the transmitter's RF output frequency is the same as the VFO frequency (i.e. no frequency multiplication between the VFO and the RF output stage). Modulation can also cause FM-ing of the VFO output if the DC voltages supplied to the VFO are also being modulated (i.e. if they are not well enough regulated).

To check this you might compare the behavior of the counter (when you modulate) using a crystal, to the behavior of the counter (when you modulate) using the VFO. If the use of a crystal significantly reduces the effect of modulation on the behavior of the counter, then the counter is responding to fluctuations of the phase and/or the frequency of the VFO's output.

If you have significant phase or frequency modulation of the VFO's output, then it won't help to change the pickoff point for the counter.

Inside the TX-1, the output frequency of the VFO is multiplied by several frequency multiplier stages to obtain the final carrier frequency. The number of multiplier stages depends on the band you select to operate on.

A good place to pick off a signal, as you suggested in your post, would be at the plate of the driver tube (V2 5763), on the driver side of the coupling capacitor (C36 200pF) that connects (via the driver's output tuned circuit) the plate of the driver tube to the grids of the 6146 output tubes. However the signal you pick off will probably contain some residual signal from the plates of the output tubes... particularly because the other side of that capacitor is where the neutralizing stub is connected...  and there may be some residual effect on the counter when you talk.

Another good place to pick off the signal, that might work well in the case of a TX-1, would be across C42 (47pF), from the grids of the 6146s to ground. If you use this pick off location (instead of the plate of the driver tube)... use the same method as described below for the case where you locate the pick off at the plate of the driver tube.

The output of the driver tube is a high impedance point, but it is a much larger signal than you would want to put into your counter.

The way to create the pickoff is the connect one end of a 0.001 uF (1000V rated) capacitor to the plate of the driver tube... and the other end of the capacitor to one end of a 50k ohm (2W or larger) non-inductive resistor*. The other end of the 50k ohm resistor is connected to one end of a 50 ohm (1/4 watt or larger) non inductive resistor. Finally, the other end of the 50 ohm resistor is connected to ground. At 3.885MHz, this capacitor and the two resistors form an approximately 1000:1 RF voltage divider, that also blocks the 370VDC on the plate of the driver tube. At 30MHz, (because of parasitic capacitance across the 50k ohm resistor) the voltage divider is roughly 100:1. The output to the counter is via a coaxial cable connected across the 50 ohm non inductive resistor of the voltage divider.

*You want the parasitic capacitance across this 50k ohm resistor to be 1pF or less. A good approach would be to use 5 non-inductive 1/2 watt resistors in series; each with a value of 10k ohms.

If the RF component of the driver's plate voltage is (roughly) 350V in amplitude, then the RF voltage delivered to the counter by the pickoff (at 3.885MHz) will be approximately 0.35V in amplitude.

There is an alternative solution that might work if your counter has an adjustable zero crossing threshold. This is the method I use. I set the threshold away from zero, and the reading on the counter is not affected by modulation of the carrier (as long as there is no FM accompanying the AM).

Stu

[K4RT]

Stu,

Thank you for taking time to work up a solution tailored specifically for the TX-1.

I compared VFO with crystal transmissions and get the same results.  I believe now that it's part of a larger RFI issue, which will require suppression.

That being said, I think an RF pickoff point to feed the counter directly would be useful.  Initially, I thought the driver tube plate might be the best place for a pick off, but I like your idea of tapping the signal across C42.  There's a little more room under the chassis there to work, and I think I can add the pickoff circuit on a large terminal strip.  I have attached an image of a schematic I believe represents your idea, although to keep the drawing simple the five non-inductive 10K resistors are shown in the drawing as a single 50K resistance. As you suggest, I would use coaxial cable across the 50 ohm resistor to the counter.

I'll let you know how it works out.

73,
Brad

[AB2EZ]

Brad

Okay... you are welcome!

I was thinking a bit more about using C42 as the pick-up point.

Since the signal level is smaller there (vs. the output of the driver), I suggest that you use a 5k ohm resistor (instead of 50k). This will produce a 100:1 voltage divider (instead of a 1000:1 voltage divider).

Since the existing 47pF capacitor has an impedance of less than 1000 ohms at all frequencies above 3.5MHz, the added 5000 resistance in parallel it will not affect the behavior of the TX-1.

A single 1/2 watt non-inductive resistor will be okay, since the average power it will have to dissipate is roughly 0.1  watts (i.e. 0.5 x 33V x 33V /5,000 ohms for a sine wave whose amplitude is 33V).

With a 33V (amplitude) sine wave across the capacitor, the output to the counter will be 0.33V... at RF frequencies up to 30MHz... if the parasitic capacitance across the 5000 ohm resistor is 1pF or less.

Stu

[K4RT]

Stu,

Thank you! I have noted the resistor value change and updated the circuit schematic here for future reference.

I forgot to mention that the approximate third of a volt output is perfect for the input of this counter (perhaps that's a standard input level for frequency counters).

Brad

[K4RT]

Stu - Thank you.  I installed your RF pickoff circuit in my Apache.  The arrow in the attached photo points to the terminal strip to which the components, leads, and shielded cable are soldered.  The strip is secured to the chassis with two screws. The shielded cable is secured with a nylon tie.

[MikeKE0ZUinkcmo]

I made a RF sampler and use it for all the various BAs.   I use a patch panel for configuring the various Exciters/Amps and put the sampler's connections on the panel as well.   Just put it between the driver and amp, or the amp out and tuner input.   The "tap" goes over to the TE bench where the counter is.

[WQ9E]

Mike,

Beautiful construction!

Has a spec of dust ever had the nerve to invade your radio room? 

[MikeKE0ZUinkcmo]

Rodger,
Oh yes, but I only clean whats gonna be in the picture.

[KA2DZT]

Rodger,
Oh yes, but I only clean whats gonna be in the picture.

That's why I never take pictures.

Fred