Thanks Roger and Roy. I think I'm starting to get it. As the data
rate gets higher the tank becomes more of an issue. The PSK31
discussion really opened my eyes.

73
Gary
W4AF


"Roger Leone" wrote in message ...
I still have trouble visualizing how a 180 or 270 degree change can occur
in a
single rf cycle and be able to overcome the "inertia" (probably a
poor choice of words) of the rf circuits , feed line and antenna
system.

Gary
W4AF


Gary:

For the phase shift to occur during one RF cycle, wouldn't that suggest that
the modulating frequency is close or equal to the RF carrier frequency? I
don't think that is the situation you are trying to visualize. The
modulating frequency, in voice or common digital modes, is more likely a
tiny fraction of the RF carrier frequency. The phase shift of the RF
carrier only has to occur at the modulating freqency, not at the RF
frequency. That means that during the modulating phase shift, many
thousands or even millions of RF cycles can occur.

If you are thinking of a mode like PSK 31, in which modulation is by phase
shift, remember that the phase shift occurs in the audio tone that is
modulating the RF signal. The RF signal can follow this phase change
easily, since many millions of RF cycles occur during the audio phase shift.

Roger K6XQ

I still have trouble visualizing how a 180 or 270 degree change can occur
in a
single rf cycle and be able to overcome the "inertia" (probably a
poor choice of words) of the rf circuits , feed line and antenna
system.

A sudden phase shift like that (within one RF cycle) only happens if you phase
modulate the carrier with a square wave, but then the tank circuit will even
things out, but will still be quite a wide bandwidth.

You don't modulate any carrier in any mode using square waves to be honest -
unless you pass that carrier thru a good filter after modulating it and before
letting it loose. You should low-pass filter the modulating signal before
letting it modulate the carrier - that way the phase change will be slow and not
sudden as you were thinking.

Clive

I still have trouble visualizing how a 180 or 270 degree change can occur
in a
single rf cycle and be able to overcome the "inertia" (probably a
poor choice of words) of the rf circuits , feed line and antenna
system.

A sudden phase shift like that (within one RF cycle) only happens if you phase
modulate the carrier with a square wave, but then the tank circuit will even
things out, but will still be quite a wide bandwidth.

You don't modulate any carrier in any mode using square waves to be honest -
unless you pass that carrier thru a good filter after modulating it and before
letting it loose. You should low-pass filter the modulating signal before
letting it modulate the carrier - that way the phase change will be slow and not
sudden as you were thinking.

Clive

Think of the phase shift as being a frequency shift (frequency can't vary
unless the phase shifts somewhere along the line. Phase modulation and
frequency modulation appear identical at the receive end.
(broke=not working, retired=not working, retired=broke)

Think of the phase shift as being a frequency shift (frequency can't vary
unless the phase shifts somewhere along the line. Phase modulation and
frequency modulation appear identical at the receive end.
(broke=not working, retired=not working, retired=broke)

(gary) wrote in message . com...
Can someone explain to me how a rf carrier that is phase shift
modulated, for lets say digital transmission, can be amplified in an
amp with a tank circuit output. Why doesnt the tank "flywheel effect"
prohibit or inhibit the abrupt phase changes in the signal. thanks

The ensuing discussion reminded me of when I learned about required
tank circuit bandwidths for some given modulation. Can you say, "hit
myself over the head with a hammer"? In this particular case, I
wanted a quick output pulse at 150MHz, and my first trial was to pulse
a lower-power stage in a multiplier chain. The output was anything
but a fast pulse! But it was still fairly wide bandwidth, compared
with what's needed in voice-bandwidth systems or probably even video.
The solution, after I realized what was going on (didn't take long...a
lot shorter time than it took to build the thing, for sure!), was to
pulse-modulate the screen of the power amplifier stage. Then the main
limitation to RF pulse rise time was the Q of the tank and the ability
of the power tube to drive that tank with high current pulses to
quickly build up the RF level.

So as others have noted, you WON'T change the phase 180 degrees in
half a cycle of the RF...it will take lots of cycles...but that's OK
because you don't NEED to change it that fast. For example, if your
digital bits last for, say, 1000 cycles of the RF, that would be a bit
time of 286 microseconds, or 3500 baud, at 3.5MHz carrier frequency.

Please note that at the receive end, you will do FAR MORE RF filtering
than on the transmitter end (in the RF tank circuits)! In fact, on
the transmitter end, you better band-limit the modulation signal
somehow, because the RF tank circuits will NOT limit it enough, in
general. So there will be filters in the modulation input path that
make SURE that you don't even try to change the phase that quickly
anyway. That's because such abrupt changes -- in phase or in
amplitude -- result in wide transmitted bandwitdh and cause
interference to other users of the spectrum. So the whole system,
including the modulation technique, data rate, transmitter, receiver,
and decoder/demodulator, work together to transmit the information in
"just enough" bandwidth to get the job done, and in general they DO
NOT use abrupt changes in phase or amplitude or frequency.

Cheers,
Tom

(gary) wrote in message . com...
Can someone explain to me how a rf carrier that is phase shift
modulated, for lets say digital transmission, can be amplified in an
amp with a tank circuit output. Why doesnt the tank "flywheel effect"
prohibit or inhibit the abrupt phase changes in the signal. thanks

The ensuing discussion reminded me of when I learned about required
tank circuit bandwidths for some given modulation. Can you say, "hit
myself over the head with a hammer"? In this particular case, I
wanted a quick output pulse at 150MHz, and my first trial was to pulse
a lower-power stage in a multiplier chain. The output was anything
but a fast pulse! But it was still fairly wide bandwidth, compared
with what's needed in voice-bandwidth systems or probably even video.
The solution, after I realized what was going on (didn't take long...a
lot shorter time than it took to build the thing, for sure!), was to
pulse-modulate the screen of the power amplifier stage. Then the main
limitation to RF pulse rise time was the Q of the tank and the ability
of the power tube to drive that tank with high current pulses to
quickly build up the RF level.

So as others have noted, you WON'T change the phase 180 degrees in
half a cycle of the RF...it will take lots of cycles...but that's OK
because you don't NEED to change it that fast. For example, if your
digital bits last for, say, 1000 cycles of the RF, that would be a bit
time of 286 microseconds, or 3500 baud, at 3.5MHz carrier frequency.

Please note that at the receive end, you will do FAR MORE RF filtering
than on the transmitter end (in the RF tank circuits)! In fact, on
the transmitter end, you better band-limit the modulation signal
somehow, because the RF tank circuits will NOT limit it enough, in
general. So there will be filters in the modulation input path that
make SURE that you don't even try to change the phase that quickly
anyway. That's because such abrupt changes -- in phase or in
amplitude -- result in wide transmitted bandwitdh and cause
interference to other users of the spectrum. So the whole system,
including the modulation technique, data rate, transmitter, receiver,
and decoder/demodulator, work together to transmit the information in
"just enough" bandwidth to get the job done, and in general they DO
NOT use abrupt changes in phase or amplitude or frequency.

Cheers,
Tom