Heathkit SB and HW series frequency meter and CAT signaler

Introduction:

The objective of this project is allowing the Heathkit SB or HW transciever to send the current frequency to a computer, in order to use n1mm or other log programs.

Preliminary considerations.

The current frequency on the HW and SB series radios is calculated by subtracting the LMO and BFO frequencies from the HFO frequency.

f_{final}=f_{hfo}-f_{bfo}-f_{lmo}

I will use a microcontroller to do the counting, math and ci-v processing.

The microcontroller for the first testing is a PIC18F2520. This will present us with some challenges that I plan to overcome either with some logic or a different PIC in the future namely:

Frequencies being read, this pic will not read more than 1.7 MHz directly, so, I’ll need a divisor.
Number of external timers, only one is available, so I’ll multiplex the readings. (or use a different pic)
Input sine wave, must be converted to a square wave to be properly read.

Bellow is a concept of the project.

Measurements:

To pick up the signals I’ll follow the instructions on the SB-650 user and assembly guide.

BFO:

The BFO signal is picked at the IF board placing a 220pF cap on point A of the picture.

The signal at this point in the radio, before the 220pF cap has 5.12 Vpp with a bias of around 2V, this will be removed by the cap.

After the said cap the signal has 4.84Vpp with almost no bias.

This signal will be enough for the circuit, but I’ll place some kind of protection before feeding it into the logic circuits.

HFO:

This signal is picked up in the band pass circuit after making a cut on one of the tracks and placing a 51Ohm resistor (I placed 2 100Ohm in parallel.)

The challenge here is the signal amplitude of around 1 to 2 Vpp. This changes with the band and is somewhat tunable. But I want to make this as adaptable as possible to various configurations that may appear on other radios.

LMO/VFO:

Picked up on the band pass board where the connector wire of the LMO attaches.

Signal here is also bellow 2 Vpp so it will need to be amplified…

CI-V Proof of concept.

To test the feasibility of the CI-V communication, I programmed a simple response loop where the program (n1mm) send a frequency request and the program replies with the current frequency.

Bellow in the picture, channel 0 has the CI-V request and channel 1 replies with the debug messages and in the last frame the CI-V reply.

Signal Amplification

To make the counter work well all signals to be measured should be at TTL levels. I’m toying with 2 options:
1) Transistor + schimtt trigger
2) Comparator, like LM311

Simulating with ltspice, I got a pretty good result with the LM311, note the bias on the input.

The Transistor route is also worth exploring in case of difficulty with the LM.

I’ll put these in practice soon.

First Conceptual Tests

They didnt Work…

Started to build the prototype today, some parts are ordered but have not arrived yest, so im using some lower performance parts that hurt the ability to cope with higher frequencies.

The concept is simple to understand but somewhat chalanging to execute because of the high frequencies and the prototyping techniques I have available.

So the signals are schimt trigered by a 74AC14 to turn the sinewave aceptable to be processed by the following chips.

After that and since we have 3 signals and 1 counter we multiplex it with a 74151, this multiplexing is controlled by the pic on the 2 LSB of the input selectors.

Then the signal is divided by 16 to be able to be counted by the PIC timer reliably.

Here is the implementation or first prototype.

With a sqare wave I can measure up to 5.5MHz, I assume that part of that results of my 7414 being a HC , ordered som ACs to check.

Unfortunatly with a sine wave only 2,4MHz can be achieved…

Next steps will involve an aplification and ‘cleaning’ of the signal to be measured.

Connecting to the pc

The project advanced again, now, I can read the signals, and send them to CAT so i can use it on N1MM or other cat sw.

The signals, BFO, LMO and HFO are fed to a 2n3904 transistor that amplifies them and feeds 3 ports of the 7414, this is the triggered signals are then sent to the 74151 mux and fed to a divider, 74939, finally the signal is fed to a pic.