Hi,

I am working on building a VCO for FSK with 850 Hz separation between
mark and space. I am building it using a 16.000 MHz 32pf parallel load
crystal. I am pulling the crystal via a varactor to implement the FSK.
However, the amplitude between mark and space is an unacceptable 10 dB
with about 700 Hz of shift. Any ideas on how to pull the crystal
without changing the amplitude of the mark and space frequencies
considerably?

A bit about my circuit: have an 8 - 50 pF tuning cap across pins 6 and
7; a 22pF cap shunts pin 7. The crystal shunts pin 6. A 100nF DC block
cap separates pin 6 from the 8 - 29 pF (3 to 10 V reverse bias)
varactor. A 1mH inductor isolates AC from the varactor biasing.

I tried adding 1uH of inductance in series with the crystal. This moved
the oscillation frequency up, but the amplitude problem remained. I
tried putting the varactor across the tuning cap. This configuration
exhibited the same amplitude shift problem, but shifted the frequency
up rather than down. I tried combining the two shifts via two varactors
and a JFET to invert the biases, one across the crystal (as originally)
and one across the tuning cap. This didn't work well (very noisy
spectrum).

Any ideas as to what could be the problem and how to fix it?

Cheers,
Chris

Chris, when you capacitively "pull" an Xtal, you lower the Q of the
tuned circuit hence the amplitude of oscillation varies with the
capacitive loading on the Xtal.

This is not so much a problem as a natural effect that requires
compensation.

Arguably the easiest way to compensate for this is to simply drive a
switching transistor or IC into saturation throughout the range of
amplitude variations that the oscillator experiences. Essentially,
you're buffering using a quasi Schmidt Trigger arrangement, which
produces a constant amplitude output.

I believe that the problem that you'll eventually face is how to keep
your 850-Hz shift constant and centered. When I was active in FSK TTY,
my solution was to incorporate a monitor scope with filters into my
VCO, plus provide real-time adjustable shift and centering adjustments.

Hope this helps.

Harry C. (ex K2JEZ)

Chris, here's a question for you. (I'm assuming that you are working
with FSK TTY.)

Is the classic 850-Hz shift still the current standard as it was during
the 1960s when I was still involved in TTY? I had thought that it had
been reduced to something like 75-Hz during the 70s due to the problems
with propagation path differences over a 850-Hz frequency spread.

Also, back in the mid-60s, it was sort of presumed that AFSK would soon
replace FSK... Did that ever actually happen?

By the way, in those days we tested our terminal units by printing NSS
transmissions for hour after hour. You could then visually scan down
the pages of printed code groups to determine how often your system
produced a glitch.

I was then among the more fortunate, because I owned a Model 26 TTY, a
model 14 Transmitter/Distributor, plus the holy of all holys, a Model
14 typing reperforator. Today all of these have been replaced by
computers! :-) Still, hearing those old machines hammer and click
away provided much more gratification than does a computer!

Have fun, Harry C.

On 5 Nov 2005 05:16:02 -0800, "Apparatus"
wrote:

Hi,

I am working on building a VCO for FSK with 850 Hz separation between
mark and space. I am building it using a 16.000 MHz 32pf parallel load
crystal. I am pulling the crystal via a varactor to implement the FSK.
However, the amplitude between mark and space is an unacceptable 10 dB
with about 700 Hz of shift. Any ideas on how to pull the crystal
without changing the amplitude of the mark and space frequencies
considerably?

A bit about my circuit: have an 8 - 50 pF tuning cap across pins 6 and
7; a 22pF cap shunts pin 7. The crystal shunts pin 6. A 100nF DC block
cap separates pin 6 from the 8 - 29 pF (3 to 10 V reverse bias)
varactor. A 1mH inductor isolates AC from the varactor biasing.

I tried adding 1uH of inductance in series with the crystal. This moved
the oscillation frequency up, but the amplitude problem remained. I
tried putting the varactor across the tuning cap. This configuration
exhibited the same amplitude shift problem, but shifted the frequency
up rather than down. I tried combining the two shifts via two varactors
and a JFET to invert the biases, one across the crystal (as originally)
and one across the tuning cap. This didn't work well (very noisy
spectrum).

Any ideas as to what could be the problem and how to fix it?

Cheers,
Chris

I'd look at your circuit. Thats a very large change for a small shift
at 16mhz. Nominally I'd expect only a tiny change at 16mhz and no
where near 10db. From what you say it sould like your using
a NE/SE602 or similar as the osc and it may be the constants
you are using are off. it also may be that the varactor you using
is getting foward biased and that will alter the oscillartor output.

For a simple shift a series cap and a diode switch is adaquate for
a shift like that. To dial it in the cap can be variable. There are
plenty of published circuits that already do this in various CW
tranceiver designs for RX to TX shift (sidetone too).

An alternate is live with it and add a limiting stage after it
(74HC04s) to take out the amplitude change.

Allison

wrote in message
oups.com...
Chris, here's a question for you. (I'm assuming that you are working
with FSK TTY.)

Is the classic 850-Hz shift still the current standard as it was during
the 1960s when I was still involved in TTY? I had thought that it had
been reduced to something like 75-Hz during the 70s due to the problems
with propagation path differences over a 850-Hz frequency spread.

Also, back in the mid-60s, it was sort of presumed that AFSK would soon
replace FSK... Did that ever actually happen?


I don't know when the 170 hz shift took over for most use. It was being
used about 25 years ago on the low bands for most of the users when I first
started rtty.

FSK is still all that is allowed on the low bands. Today it is generated by
using a SSB transmitter in the SLB mode and feeding in audio tones. This
generates a signal that is the same comming out of the transmitter as a FSK
transmitter. That is possiable due to the way a ssb transmitter works. If
the same rig was switched to AM or FM them it would be AFSK.

YOu can go here and download some programs that you can use on the computer
to do the digital modes. As I mentioned above by feeding the audio tones
into the mic or audio input of a ssb transmitter you will be generating FSK
eventhough it might not seem so.

http://www.muenster.de/~welp/sb.htm

One of the best free programs to get started with in rtty is the MMTTY
program.

Chris,
You didn't say what you're using for th active element...you only gave pin
numbers. Can you elaborate?

Joe
W3JDR


"Apparatus" wrote in message
oups.com...
Hi,

I am working on building a VCO for FSK with 850 Hz separation between
mark and space. I am building it using a 16.000 MHz 32pf parallel load
crystal. I am pulling the crystal via a varactor to implement the FSK.
However, the amplitude between mark and space is an unacceptable 10 dB
with about 700 Hz of shift. Any ideas on how to pull the crystal
without changing the amplitude of the mark and space frequencies
considerably?

A bit about my circuit: have an 8 - 50 pF tuning cap across pins 6 and
7; a 22pF cap shunts pin 7. The crystal shunts pin 6. A 100nF DC block
cap separates pin 6 from the 8 - 29 pF (3 to 10 V reverse bias)
varactor. A 1mH inductor isolates AC from the varactor biasing.

I tried adding 1uH of inductance in series with the crystal. This moved
the oscillation frequency up, but the amplitude problem remained. I
tried putting the varactor across the tuning cap. This configuration
exhibited the same amplitude shift problem, but shifted the frequency
up rather than down. I tried combining the two shifts via two varactors
and a JFET to invert the biases, one across the crystal (as originally)
and one across the tuning cap. This didn't work well (very noisy
spectrum).

Any ideas as to what could be the problem and how to fix it?

Cheers,
Chris

Thank you all for your informative replies.

Turns out I had two problems. One was that my scope probe capacitance
(30pF) was loading the oscillation. However, the spectrum analyzer is
sensitive enough though that if I simply place the probe close to the
oscillator, the radiated signal from the oscillator is picked up.

The second problem was my bench, for I tried the antenna probe trick
before without success. I had the solder joints on my board touching
the ESD protection mat that was acting like a semi-conductive surface I
think. Once I placed the board on spacers the issue resolved itself.

I now get about 2kHz of tuning over 9V with 2.8dB of amplitude change
(this includes a slightly non-linear region of the varactor in the last
2-3 volts). Selecting an 850 Hz interval and applying a 100 Hz
modulating square wave produces a beautifully symmetric FSK spectrum.

Cheers,
Chris