Anyone have access to a VHF spectrum analyzer?

Frequency and phase modulations do not belong to what are called "linear modulations". In few words this means that the modulated signal obtained in response to a pilot tone alone and the modulated signal corresponding to a pilot tone plus something else have nothing in common.

Even the case of a carrier being frequency-modulated by _one_ sinusoidal signal is rather complex: you end up with an (theoretically) infinite number of lines computed from bessel functions. The response to _two_ sinusoidal signals is _not_ related to the prior response.

In view of this, perhaps you may reconsider what you are looking for.

Pere

The 15 kHz LPF prior to the modulator will do nothing to the RF spectrum out of the FM modulator. Commercial FM transmitters already low pass filter the audio, but that doesn't reduce the audio sideband levels after modulation.

Now if you're thinking about the baseband rather than RF, then yes - low pass filtering the audio prior to the FM modulator will prevent spillover into the 19 kHz slot occupied by the pilot. But as I mentioned, commercial transmitters do this already.

off

rox

What kind of analyzer did you use, John? I can probably get a decent picture for him when I get back to the office in January; I suppose it will be difficult to get a good spectrum of the continuously changing FM modulation with a swept analyzer, but our FFT instrumentation generally does pretty well. I didn't exactly get what he's trying to do, but we don't have any trouble demodulating FM digitally on signals that aren't "HEAVILY amplitude limited".

Cheers, Tom

It's an Aeroflex/IFR 2399C, 3 GHz. It uses FFT for narrowband sweeps, but not for the wider stuff. I must have been on a music station, because the spectrum was all over the place from sweep to sweep, not a lot of structure except for the two rectangular plateaus high and low. I couldn't resolve any stereo carrier stuff, but I didn't try really hard.

John

It isn't transmitted. That's why the 19 kHz pilot signal is. It is doubled and used in place of the missing 38 kHz carrier.

--
http://improve-usenet.org/index.html

aioe.org, Goggle Groups, and Web TV users must request to be white
listed, or I will not see your messages.

If you have broadband, your ISP may have a NNTP news server included in
your account: http://www.usenettools.net/ISP.htm


There are two kinds of people on this earth:
The crazy, and the insane.
The first sign of insanity is denying that you\'re crazy.

Nice of you to do that for the op.

Ed

What about the stereo difference signal / pilot tone and SCA subcarrier; all of which are above the 15 kHz audio filter cutoff and present in the modulating signal.

The low pass filter is in the mono audio path to prevent interference with the 19 KHz pilot. Any filter at the transmitters would be bypassed with a switch or relay for stereo or SCA applications. Just like TV transmitters that had a chroma trap on the video input that had to be switched out to use it for color broadcasting. If it was switched out of the circuit, enough noise could get through to prevent the color killer circuit from working in the TV sets..

--
http://improve-usenet.org/index.html

aioe.org, Goggle Groups, and Web TV users must request to be white
listed, or I will not see your messages.

If you have broadband, your ISP may have a NNTP news server included in
your account: http://www.usenettools.net/ISP.htm


There are two kinds of people on this earth:
The crazy, and the insane.
The first sign of insanity is denying that you\'re crazy.

No, they don't. The audio program does not frequency modulate the subcarrier; it is added to together linearly.

Those are mixed in with the low-pass-filtered audio signal (after the low-pass step) and so they're part of the total signal fed to the FM modulator. As you point out, they're all above the 15 kHz cutoff.

Since they're part of the modulation, they will result in the creation of FM sidebands which lie on either side of the carrier (and they'll be more than 15kHz away). For example, the pilot tone will create sidebands at 19, 38, 57, ... away from the carrier, while the 38 kHz stereo subcarrier will result in a primary sideband at a 38 kHz offset from the RF carrier frequency.

There will, in principle, be additional sidebands out at 19 and 38 kHz multiples. However, the strength of these additional sidebands will be very low, because the pilot-tone and subcarrier signals are of low amplitude and aren't modulating the carrier strongly. I imagine that the subcarrier level is deliberately limited in order to keep the higher-order sidebands from "splattering" out of the 150 kHz RF passband strongly enough to interfere with the next station's signal.

--
Dave Platt                                    AE6EO
Friends of Jade Warrior home page:  http://www.radagast.org/jade-warrior
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!

Those are inserted after the low-pass-filtered main audio channel. They are present as you noted in the modulated signal.

I have made my point.

w

er

ercial

=A0 =A0 =A0 =A0 =A0 =A0 =A0 AE6EO

r

Um, the 38kHz subcarrier is suppressed. For all practical purposes it isn't there.*

There are standards for the 19kHz pilot tone level: it's supposed to be between 8% and 10% modulation. Assuming 10%, that's a modulation index of .394, if I didn't slip any cogs. If you catch it at a time when there's no other modulation, and the unmodulated carrier is 1 amplitude unit, that modulation index yields a carrier amplitude of

0.961, and sideband amplitudes (in order) of .193, .019, .0012 -- and stuff down more than 80dB. Because of intermodulation among all the frequency components of a complex modulating signal, things get messy when they're talking or playing music.

Cheers, Tom

*Specifically the subcarrier must be suppressed to less than 1% deviation, or a modulation index of .02. The first sideband will be more than 40dB below the unmodulated carrier, and the second sideband more than 86dB below. I suspect modern commercial broadcast equipment has quite a bit better subcarrier suppression than that upper limit.

io

he

nd

y

of

e.

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 AE6EO

out

r the

.

cial

Just to amplify this a little... It's important to realize that the _modulation_ and thus a perfectly demodulated signal will have nice bands: 0-15kHz baseband, a 19kHz pilot tone "in the clear," a notch right at 38kHz (since the [L-R] information doesn't go all the way to DC), and the [L-R] double-sideband suppressed carrier information from

38-15kHz to 38+15kHz (23kHz to 53kHz). Of course the 15kHz cutoffs aren't brick-wall.

BUT--the spectrum of the FM modulated signal will have significant energy from carrier-75kHz to carrier+75kHz, nominally. There will NOT be nice clean little areas at carrier +/- 19kHz for the pilot tone to sit in. The whole spectrum will be filled in by the vast multitude of sidebands generated by the complex modulating signal, and by intermodulation among those sidebands (which is a result of perfect FM modulation, not any transmitter non-linearity). At some moments, you may find a relatively simple spectrum, such as when a single musical instrument such as a flute is playing a sustained clean note, but in general the material that's broadcast is far more complex than that. (I recall looking at the spectrum of a note held by for a relatively long time by one particular singer, and was amazed to find it was sinusoidal with only about 10% harmonic distortion. That was unusual!)

Cheers, Tom

Um, the 38kHz subcarrier is suppressed. For all practical purposes it isn't there.*

There are standards for the 19kHz pilot tone level: it's supposed to be between 8% and 10% modulation. Assuming 10%, that's a modulation index of .394, if I didn't slip any cogs. If you catch it at a time when there's no other modulation, and the unmodulated carrier is 1 amplitude unit, that modulation index yields a carrier amplitude of

0.961, and sideband amplitudes (in order) of .193, .019, .0012 -- and stuff down more than 80dB. Because of intermodulation among all the frequency components of a complex modulating signal, things get messy when they're talking or playing music.

Cheers, Tom

*Specifically the subcarrier must be suppressed to less than 1% deviation, or a modulation index of .02. The first sideband will be more than 40dB below the unmodulated carrier, and the second sideband more than 86dB below. I suspect modern commercial broadcast equipment has quite a bit better subcarrier suppression than that upper limit.

Right. The 38kHz tone is not transmitted. The receiver re-creates the 38 kHz by doubling the received 19 kHz pilot, and uses it for detection of the upper stereo audio bands. Nothing is lost by doing this. In fact, the transmitter's 38 kHz was generated by doubling the 19 kHz.

In older equipment. Since about 1968 they started using 76 kHz PLL and a divider chain to 38 kHz and 19 kHz.

ff

ox

Now that I'm back in the office, the measurement is easy, though not very useful. As I suggested in another posting in this thread, almost all the time (e.g., when someone's talking or music is playing) there's energy practically continuously from carrier minus a few tens of kHz to carrier plus a few tens of kHz. Typical shape on a linear- in-dB display is roughly triangular to roughly Gaussian, with the peak at the carrier. When there's "dead air" (no audio), the 19kHz pilot tone sidebands are easily visible, but all stations do a good job of avoiding dead air.

Note that these spectra of FM broadcast signals are much different from the spectra of the old NTSC television stations or cable NTSC signals, which are amplitude modulated. In those, you can very easily pick out energy at harmonics of the vertical scan rate and of the horizontal scan rate. You can easily see the color subcarrier as a very narrow spike. You can easily see the frequency-domain interleaving of the dominant horizontal scan rate sidebands about the color subcarrier between the horizontal scan rate baseband sidebands: the folks that developed that system back in the early days of color TV understood quite a lot about the spectral content of the signals they were dealing with (and a whole lot of other things).

A whole lot of FM stations in this market also transmit a digital signal as well, appearing as sideband bands with practically uniform energy density over a cleanly limited band of frequencies.

The measurements were taken with a very fast FFT-based system, using

70Hz resolution bandwidth. With narrow enough bandwidth you _might_ be able to reliably see the 19kHz sidebands in the presence of other modulation, but that seems a really dumb way to try to do it. Much better to demodulate the signal first.

Cheers, Tom