[CW] Good Sounding Sidetone or Code Practice Oscillator

[Donald Chester]

I am finally back on CW after almost a year. My sidetone oscillator crapped
out, and I can't send very well with the bug by feel if I don't have a
sidetone monitor. I finally got it going this weekend. One thing it taught
me is that my limits are about reached troubleshooting electronics at circa
1980s technology.
 
The thing is sort of a homebrew conglomeration using the  carcass of  a HAL
2550 electronic keyer, the electronic keyer board from a the Yaesu FT-901,
and a homebrew cathode keying circuit copied from the 1981 ARRL Handbook. I
picked up the HAL at a hamfest in the early 80s. It was supposed to be a
fancy top-of-the-line keyer, but I didn't like the way it sent code.
Something was odd about the sequencing and timing seemed off, or at least I
never could get it  to work right for me. I didn't have a  schematic or
manual, so to me the whole thing was just a jumbled up mess of ICs mounted
on a circuit board and I didn't have a clue what anything was, except for
the power supply components.
 
So I took the keyer board from a scrapped FT-901. It uses one of those
little Curtis iambic keyer ICs. Then I built up an electronic cathode keying
circuit as described in the 1981 ARRL handbook, using a small transistor
driving  the biggest TV sweep transistor I could find, one rated at about
1400 volts and several amps, to serve as an electronic relay to key the
cathode of any of my tube type transmitters. The sweep transistor  circuit
also operates from a  hand key, drawing only a couple of milliamps, so it
doesn't burn up the contacts on my bug trying to cathode key the driver
stages in my transmitters, which draw between 40 and 200 ma depending on
which rig.

I used the power supply and enclosure from the Hal unit, and mounted the
FT-901 keyer board over the main circuit board, disconnected all the HAL
circuitry, and wired all the inputs and outputs to the appropriate original
jacks on the enclosure.  I was able to tap the key line at the output of the
Yaesu keyer board, to activate the original sidetone oscillator in the HAL.
This worked fine for about 30 years, until the sidetone quit a few months
ago. Not having any info on the HAL unit, I decided to just build up a
separate sidetone oscillator, using a 12AU7, based roughly on the "Little
Oskie" described in a 1955 QST. I must have spent a week getting the thing
to work using parts salvaged from a junked signal generator. It makes a fine
code practice oscillator, but the shaping circuitry and key click filters in
the transmitters give the sidetone a "whoop-whoop" sound, so I decided that
was a dead end. 
 
I found a downloadable manual to the HAL unit at the BAMA website, with
schematic. First of all, the power supply didn't work properly, too much
voltage sag under load, and the +5v  regulated output showed zero volts. I
found the voltage sag problem was ONE DIODE in the bridge rectifier, so that
was easily fixed. Then, instead of trying to troubleshoot the voltage
regulator circuit, I looked in my junkbox and found one of those little
3-leg 5-volt one amp voltage regulator chips, and tore out all the HAL
voltage regulator components and substituted the 5v regulator, getting rid
of a hand full of transistors, resistors and capacitors. But the  sidetone
still wouldn't work, even though I now had all the right voltages out of the
power supply.
 
I started probing around in the sidetone oscillator circuit and found one
bad  diode. Replacing that still didn't make it work, so I tested the IC
that has a bunch of NAND gates driving another IC which is some kind of
"timing" chip to generate the tone. I looked up the reference diagram for
the NAND gate IC and checked voltages per the "truth" table. It didn't give
the proper highs and lows with the given inputs, so I looked through a
drawer full of random ICs accumulated in my junk collection, and found an
exact replacement chip. Taking the old IC off the board was a challenge, but
I managed to get the new IC in, and the thing worked per the truth table.
Re-connected a couple of circuit board traces I had cut while testing, and
the sidetone was back! I was glad I kept the old 1980 ECG solid state
devices reference book from back when I worked in a 2-way radio  shop many
years ago.
 
The problem seemed to be multiple failures that must have happened all at
once. Lightning surge, maybe. It's really frustrating to try to fix multiple
problems in the same unit, even something as simple as that sidetone
oscillator, when you don't know where to start because nothing works. It
brought back old memories of what it was like in the radio shop when I
worked  with a lot of solid state equipment using discrete components. I
don't think I could handle modern  day SMT and 100-pin ICs. I got out of
electronics as a profession just in time, back in 1984 when I started the
teaching gig. Most of my electronics work since then has been with my ham
station, which remains 90% tube technology.


Don k4kyv