Class C operation seems very well suited to HF CW operation. At least
if the resulting envelope looks "nice" by ARRL handbook definition of
keyclick reduction, I guess.

But is it good enough for RTTY?

How about for PSK-31? Olivia?

What I've got is an old Eico 720 (Novice 75W-class transmitter biased
for deep class C, for those too young to remember) that I currently
use for CW, but would like to try out for RTTY or maybe PSK-31 by
feeding appropriate waveforms into the "VFO in" jack. My gut feeling
is that RTTY would be OK but don't really understand how class C will
mess up the PSK-31 waveform or introduce intermod that will make the
resulting spectrum broader than necessary.

The Class C stages will clearly fail the usual SSB two-tone testing
but what is a more appropriate test to say that it's good enough - or
not - for PSK-31?

Tim N3QE

On Mar 11, 6:41*am, Tim Shoppa wrote:
Class C operation seems very well suited to HF CW operation. At least
if the resulting envelope looks "nice" by ARRL handbook definition of
keyclick reduction, I guess.

But is it good enough for RTTY?

How about for PSK-31? Olivia?

What I've got is an old Eico 720 (Novice 75W-class transmitter biased
for deep class C, for those too young to remember) that I currently
use for CW, but would like to try out for RTTY or maybe PSK-31 by
feeding appropriate waveforms into the "VFO in" jack. My gut feeling
is that RTTY would be OK but don't really understand how class C will
mess up the PSK-31 waveform or introduce intermod that will make the
resulting spectrum broader than necessary.

The Class C stages will clearly fail the usual SSB two-tone testing
but what is a more appropriate test to say that it's good enough - or
not - for PSK-31?

Tim N3QE

PSK-31 is just slow phase-shift keying, no? If the 720 had very
narrow tuned circuits in it, it could be a problem preserving the
fidelity of really fast phase-shift keying, but for the speeds
involved, a very tiny percentage of the bandwidth of the transmitter,
there simply shouldn't be any problem. I suppose there could be an
issue if the transmitter tries to self-oscillate, but if it does,
you've got bigger problems than PSK-31 transmission!

Cheers,
Tom

On Wed, 11 Mar 2009 06:41:55 -0700, Tim Shoppa wrote:

Class C operation seems very well suited to HF CW operation. At least if
the resulting envelope looks "nice" by ARRL handbook definition of
keyclick reduction, I guess.

But is it good enough for RTTY?

How about for PSK-31? Olivia?

What I've got is an old Eico 720 (Novice 75W-class transmitter biased
for deep class C, for those too young to remember) that I currently use
for CW, but would like to try out for RTTY or maybe PSK-31 by feeding
appropriate waveforms into the "VFO in" jack. My gut feeling is that
RTTY would be OK but don't really understand how class C will mess up
the PSK-31 waveform or introduce intermod that will make the resulting
spectrum broader than necessary.

The Class C stages will clearly fail the usual SSB two-tone testing but
what is a more appropriate test to say that it's good enough - or not -
for PSK-31?

Tim N3QE

If the signal has a constant envelope then it'll be fine with class C.

RTTY should be fine, I don't know about PSK-31 or Olivia.

--
http://www.wescottdesign.com

On Wed, 11 Mar 2009 11:55:11 -0700, K7ITM wrote:

On Mar 11, 6:41Â*am, Tim Shoppa wrote:
Class C operation seems very well suited to HF CW operation. At least
if the resulting envelope looks "nice" by ARRL handbook definition of
keyclick reduction, I guess.

But is it good enough for RTTY?

How about for PSK-31? Olivia?

What I've got is an old Eico 720 (Novice 75W-class transmitter biased
for deep class C, for those too young to remember) that I currently use
for CW, but would like to try out for RTTY or maybe PSK-31 by feeding
appropriate waveforms into the "VFO in" jack. My gut feeling is that
RTTY would be OK but don't really understand how class C will mess up
the PSK-31 waveform or introduce intermod that will make the resulting
spectrum broader than necessary.

The Class C stages will clearly fail the usual SSB two-tone testing but
what is a more appropriate test to say that it's good enough - or not -
for PSK-31?

Tim N3QE

PSK-31 is just slow phase-shift keying, no? If the 720 had very narrow
tuned circuits in it, it could be a problem preserving the fidelity of
really fast phase-shift keying, but for the speeds involved, a very tiny
percentage of the bandwidth of the transmitter, there simply shouldn't
be any problem. I suppose there could be an issue if the transmitter
tries to self-oscillate, but if it does, you've got bigger problems than
PSK-31 transmission!

PSK31 uses a raised cosine amplitude weighted amplitude function in order
to contain the occupied bandwidth. Use of a class C amplifier disrupts
the weighting and causes the bandwidth to spread considerably. The
demodulator will still work fairly well but the spectral spreading is
considered un-neighborly!

bart
wb6hqk

On Wed, 11 Mar 2009 21:11:18 -0500, Bart Rowlett wrote:

On Wed, 11 Mar 2009 11:55:11 -0700, K7ITM wrote:

On Mar 11, 6:41Â*am, Tim Shoppa wrote:
Class C operation seems very well suited to HF CW operation. At least
if the resulting envelope looks "nice" by ARRL handbook definition of
keyclick reduction, I guess.

But is it good enough for RTTY?

How about for PSK-31? Olivia?

What I've got is an old Eico 720 (Novice 75W-class transmitter biased
for deep class C, for those too young to remember) that I currently
use for CW, but would like to try out for RTTY or maybe PSK-31 by
feeding appropriate waveforms into the "VFO in" jack. My gut feeling
is that RTTY would be OK but don't really understand how class C will
mess up the PSK-31 waveform or introduce intermod that will make the
resulting spectrum broader than necessary.

The Class C stages will clearly fail the usual SSB two-tone testing
but what is a more appropriate test to say that it's good enough - or
not - for PSK-31?

Tim N3QE

PSK-31 is just slow phase-shift keying, no? If the 720 had very narrow
tuned circuits in it, it could be a problem preserving the fidelity of
really fast phase-shift keying, but for the speeds involved, a very
tiny percentage of the bandwidth of the transmitter, there simply
shouldn't be any problem. I suppose there could be an issue if the
transmitter tries to self-oscillate, but if it does, you've got bigger
problems than PSK-31 transmission!

PSK31 uses a raised cosine amplitude weighted amplitude function in
order to contain the occupied bandwidth. Use of a class C amplifier
disrupts the weighting and causes the bandwidth to spread considerably.
The demodulator will still work fairly well but the spectral spreading
is considered un-neighborly!

bart
wb6hqk

So, PSK-31 is out.

(Unless you want to get fancy and amplitude modulate to preserve the
envelope at the same time that you phase modulate the VFO input to
preserve the phase. There's almost no sane reason to try this, but it'd
be fun if it worked. Kind of a "Look how well my old iron works! (pay no
attention to the DSP behind the curtain).")

--
http://www.wescottdesign.com

On Wed, 11 Mar 2009 23:36:57 -0500, Tim Wescott
wrote:

On Wed, 11 Mar 2009 21:11:18 -0500, Bart Rowlett wrote:

PSK31 uses a raised cosine amplitude weighted amplitude function in
order to contain the occupied bandwidth. Use of a class C amplifier
disrupts the weighting and causes the bandwidth to spread considerably.
The demodulator will still work fairly well but the spectral spreading
is considered un-neighborly!

bart
wb6hqk

So, PSK-31 is out.

(Unless you want to get fancy and amplitude modulate to preserve the
envelope at the same time that you phase modulate the VFO input to
preserve the phase. There's almost no sane reason to try this, but it'd
be fun if it worked. Kind of a "Look how well my old iron works! (pay no
attention to the DSP behind the curtain).")

A plate modulated AM transmitter is basically a class C amplifier with
variable supply voltage, unfortunately it is hard to get 100 %
modulation with such simple systems, i.e. it is hard to get the power
down to 0. However, much of the PSK-31 envelope can still be
simulated.

Plate modulated AM transmitters for voice communication might have a
modulation transformer with a quite limited bandwidth, which might not
work well the slow PSK31 envelope.

Paul OH3LWR

On Mar 11, 9:41*am, Tim Shoppa wrote:
Class C operation seems very well suited to HF CW operation. At least
if the resulting envelope looks "nice" by ARRL handbook definition of
keyclick reduction, I guess.

But is it good enough for RTTY?

How about for PSK-31? Olivia?

What I've got is an old Eico 720 (Novice 75W-class transmitter biased
for deep class C, for those too young to remember) that I currently
use for CW, but would like to try out for RTTY or maybe PSK-31 by
feeding appropriate waveforms into the "VFO in" jack. My gut feeling
is that RTTY would be OK but don't really understand how class C will
mess up the PSK-31 waveform or introduce intermod that will make the
resulting spectrum broader than necessary.

The Class C stages will clearly fail the usual SSB two-tone testing
but what is a more appropriate test to say that it's good enough - or
not - for PSK-31?

Tim N3QE

Its not just a matter of whether or not it will work but whether it
not it will work within the confinments of the allocatted bandwidth. I
think any mode normally used with a SSB transmition would probably
flunk this test.


Jimmie