parnass@ihu1h.UUCP (Bob Parnass, AJ9S) (02/25/85)
Pardon me if you've seen this already, but article posting
here has been buggy:
======
EAVESDROPPING ON SUBCARRIER TRANSMISSIONS
A few weeks ago, I posed a question about listening to
SCA transmissions on the FM commercial broadcast band.
An article in _M_o_n_i_t_o_r_i_n_g__T_i_m_e_s claimed that connecting
a vlf receiver to an FM broadcast receiver would permit
SCA detection.
Necessity being the mother of invention, a way was
found to confuse the ICOM R71A into tuning below 100
kHz1, and after receiving substantial inspiration from
Will Martin, Phil Karn, and others I now can report
success!
With the R71A in the FM mode, tuned to 67 kHz, I con-
nected the ICOM's vlf antenna input through a 0.1 ufd
capacitor2 to the earphone jack of a $16 General Elec-
tric AM/FM portable radio, and can now listen in on the
world of SCA!
In the first few minutes of tuning around, I've heard
the Physicians' Network, Muzak, commodity reports, and
several data transmissions.
I built a simple SCA interface, consisting of a capaci-
tor and resistor, into a plastic film canister.
+-----------------------+
| AM/FM radio |
| |
| earphone jack |
| (ring) (tip) |
+-----------------------+
| |
+--/|/|/|/--+
| |
| 100 ohm |
| 1 watt |
| |
| |
| -----
| ----- 0.1 ufd
| |
+-----------------------+
| (gnd) |
| antenna jack |
| |
| ICOM R71A |
+-----------------------+
Most activity is heard with AM/FM radio tuned to the FM
broadcast band, and the R71A set to 67 kHz FM. Aside
from the Muzak, commodity reporting, and the Physici-
ans' Network, I also heard an announcer reading from
_P_o_p_u_l_a_r__C_o_m_m_u_n_i_c_a_t_i_o_n_s on CRIS, the Chicago Radioland
Information Service. This service carries programming
of interest to the handicapped.
With the AM/FM radio tuned to the AM broadcast band,
and the R71A set to 60 kHz AM, I can hear what seems to
be stereo subcarriers on AM broadcast stations claiming
to transmit in AM stereo.
__________
1. Parnass, Bob, "Trick the ICOM R71A below 100 kHz", to be
published in _M_o_n_i_t_o_r_i_n_g__T_i_m_e_s_.
2. The capacitor is needed with the GE radio I used to
block DC.
--
===============================================================================
Bob Parnass, Bell Telephone Laboratories - ihnp4!ihu1h!parnass - (312)979-5414 jhs%Mitre-Bedford@d3unix.UUCP (03/01/85)
Ring? Not Sleeve? On an AM/FM radio? With stereo headphones? That would give out-of-phase audio in the two earpieces! de W3IKG
jhs%Mitre-Bedford@d3unix.UUCP (03/11/85)
Who, me? I will be happy to try to answer the question, although I didn't think I had anything to do with it except maybe to make a wise comment or two... " I keep feeling that this is something that I should understand-- at least qualitatively-- but I don't! I assume that one receiver is acting as a converter for the other. Please enlighten! I wonder if something similar cfeeling that this is something that I should understand-- at least qualitatively-- but I don't! I assume that one receiver is acting as a converter for the other. Please enlighten! I wonder if something similar could be done to use an F.M. receiver to receive the cordless telephones in the range 1.6 to 1.8 mhz.? TNX! Jim (W2OZH) " As I recall, the setup described by Bob Parnass (I think that was who it was) was to hook a VLF receiver to the output of an ordinary FM receiver to pick up SCCA (sic - I always call it that, Sports Car Club of America, when it should really be SCA - Subsidiary Communications Authorization or something!) transmissions multiplexed on ordinary FM broadcasts. To understand what is going on, you need to know what the FM station actually transmits. Let's do it in "top down" fashion. All you computer jocks out there should relate nicely to that. First of all, the FM station has a "composite audio" input - this is just the input on which SOMETHING (consider it a stub subroutine to be written later) is fed in to the modulator. >From the FM modulator's point of view, what you put on this input is just the MODULATING SIGNAL which you want to Frequency Modulate (FM) the station's carrier. Intuitively, you can think of the station as putting out a "pure" carrier at frequency F when this modulating signal is zero. When it is NONZERO, however, the instantaneous frequency of the transmitter is changed. Say the modulating signal value, in volts, or whatever, is M. Then the transmitter output frequency is set to F + K*M, where K is a sensitivity constant which is unimportant except that whatever maximum value of M is applied to the FM modulator input should result in a maximum "frequency deviation" K*M which is just about the maximum that the FCC allows. If this station were a plain old-fashioned monaural FM station, all you would have to do is feed in the audio signal -- voice, music or funny sound effects -- which you want the listeners out there in Radio-Landt to hear. The signal would modulate the carrier, be amplified, fed to the antenna, and radiated. It would come swooping down into your ordinary, Monaural FM receiver. The receiver would say, hmm... The frequency is F + K*M. I therefore have to subtract F and divide by K to give my loyal owner the instantaneous value of the signal waveform, which is M. Voila! Out of the FM demodulator comes M in livid high fidelity. (I don't think the following information is needed for present purposes, but I include it so somebody won't say I am oversimplifying things!!!) Now, even in simple monaural FM there is one trick that we haven't mentioned. This is called pre-emphasis and de-emphasis. Very early in the game, it was noticed that in FM systems if you fed in no modulation at all and listened to the receiver's reconstructed value of M (after it did the arithmetic noted above), there was a NOISE output from the receiver, even at fairly strong signal levels. The noise was particularly noticeable because its amplitude increased with frequency. Thus a quite noticeable high-frequency hiss was present on even fairly good signals. Somebody then had the bright idea that they should effectively "turn down the treble control" at the receiver. A fixed frequency compensating network called a "de-emphasis" network was designed and standardized to do that. But then the music had its highs "de-emphasized", so an "inverse" network called a pre-emphasis network was added at the transmitter. Next, along came stereo. In order to avoid the wrath of all the owners of monaural FM sets, the FCC in its wisdom decreed that a "compatible" system would be necessary before they would approve FM stereo. The engineers quickly noted that the A+B signal from two microphones gives a passable monaural signal (especially if "one-point" miking is used). So now the problem was how to get A and B out of A+B. Well, as your high school algebra teacher probably taught your, (A+B)+(A-B)=A and (A+B)-(A-B)=B. So all they had to do is send (A-B) in some clever way and the receiver could reconstruct A and B by "matrixing". The method adopted was to "multiplex" this (A-B) signal onto the main carrier by using it to modulate a SUBcarrier located at 38 KHz. Double-Sideband Suppressed Carrier (DSBSC) modulation was chosen. This gave a "lower sideband" extending downward from 38KHz (less 20Hz or so) to 23KHz (because the highs were cut off at 15 KHz.) A "pilot carrier" was put in at 19 KHz which allows the receiver to recover the precise frequency and phase of the 38 KHz carrier so that recovery of the (A-B) signal could proceed. In fact it turns out that this carrier can be used in an even more clever way to recover A and B signals directly from the "composite" signal ( A+B plus pilot carrier plus A-B ). Note that all the new junk -- pilot at 19KHz and A-B from 23 to 38 KHz -- are so high in frequency that most people wouldn't hear them, and most older monaural FM sets and loudspeakers won't reproduce them audibly anyway. Well, as if this wasn't bad enough, then along came the SCCA (I warned you about my warped sense of humor) and asked for authorization to put MUZAK on the air. The obvious thing to do was to put on yet another subcarrier, this time at a frequency far enough above the audio so that it wouldn't interfere with stereo broadcasting. 67 KHz was chosen as the magic frequency. But this time, FM modulation was chosen. Remember, all of this stuff is being stacked up in frequency above the normal monaural FM broadcast audio in such a way that a normal FM receiver won't be affected (much) by it. So before we even go into the "composite audio" input of our simple FM transmitter, we have a VERY "composite" signal indeed. One might diagram it like this: weak pilot suppressed just a carrier carrier boundary FM / / / sidebands / / / sub / / / / car. / | | | | | / | Normal FM | (A-B) Lower | (A-B) Upper | /||||||||\ | | audio (A+B) | Sideband | Sideband | /||||S|C|A|||||\ | | | | | /|||||S|i|g|n|a|l||||\ | DC 19KHz 38KHz 53KHz 67KHz 81KHz --monaural approx signal---- --------stereo composite signal----------- | -------------full composite (stereo + SCA) signal-------------------| Now to the main point. The FM transmitter takes this composite audio signal as input and FM MODULATES its carrier with it. When this is received by an FM receiver, what comes out of the discriminator is this composite signal. In a normal monaural FM receiver, this is fed to the volume control and thence to the audio amplifier stages, where "de-emphasis" is applied as noted above, and finally out to the speaker or headphones. The de-emphasis circuit tends to reduce the amplitude of the SCA signal since it is at a high frequency (53 to 81 KHz, with carrier at 67 KHz. However, enough gets through that it can still probably be used even at the speaker jack, since it is an FM signal and the absolute amplitude isnt too important. The best way to recover the SCA audio, however, would be to go in to the FM receiver and tap off at the "top" of the volume control -- the hot side where the discriminator signal is brought in. Bring this signal out to a jack which you can install on the set for all such nefarious purposes. In fact, make sure that the de-emphasis network is not installed ahead of this point, although in my experience it usually is not. Anyway, once you have brought out the full composite signal, the SCA signal is STILL AN FM SIGNAL that needs an FM discriminator to turn it back into normal audio. It happens to have a carrier frequency of 67 KHZ because that is what was used as the carrier at the studio when the composite audio signal was made up. So if you have a VLF receiver capable of covering 67 KHz, you can feed the composite audio signal into its input, which happens to be the antenna connection. You can then tune it to 67 KHz and switch it to FM mode and you should get perfect SCA audio out. If it doesn't have FM demodulation available at this frequency, then you can still use a trick called "slope detection". Tune it above 67 KHz so that the 67 KHz carrier falls on the lower slope if the IF selectivity curve. I.e. the signal has fallen off about halfway from its peak value on the meter. Select AM demodulation, and then tune for the best sound. You should get acceptable recovery of the audio. If there is a bandwidth switch, experiment with it. Probably the widest bandwidth available will work best. If you don't have a suitable VLF receiver covering 67 KHz, the simple trick with the Phase-Locked Loop device will work just fine. After all, the monaural FM receiver has done most of the work of picking up the weak signal at VHF (88 to 108 MHz) and amplifying it and FM demodulating it so you have the baseband composite signal available to fool around with. What is going on here is that the PLL locks to the 67 KHz signal, but since that is being FM modulated, it has to work a bit to stay locked. The Voltage-Controlled Oscillator has to be pushed above or below 67 KHz, its natural frequency, by applying a control voltage to the VCO input. All of this is done "automatically" by the design of the phase-locking circuitry, but YOU can benefit by all this work it is doing: the control voltage is precisely a measurement of the instantaneous modulation value, M, that YOU want to hear! So you pick it off, amplify it, and listen to MUZAK (yuk) to your heart's content. >From the above rather long-winded description, you can see that there is a lot of other junk floating around on the composite signal, namely all the stereo stuff below 53 KHz. To make the SCA demodulator work well and be free of interference from the main channel, it would be well to put a little selectivity ahead of the SCA demodulation if you are using the PLL trick. I haven't experimented with this, but I would think a simple high-pass filter to attenuate everything below 53 KHz would do fairly well. Of course with the VLF receiver you have all kinds of selectivity, so there should be no problem. I hope this helps you understand what is going on with SCA. If you followed all the details, it should be clear that there is no black magic going on, but that once the composite signal has been received, a VLF FM receiver (a rather rare beastie) is logically what is needed to recover the SCA signal. That's why it happens to be possible to string together such a seemingly unlikely pair of receivers and get SCA broadcasts. In the case of cordless telephones in the 1.8 MHz region, there is no such composite modulation going on. So far as I know, it's just ordinary AM on the base-to-portable link which is at that frequency. So stringing together receivers wouldn't do much for you. It IS possible, however, to MODIFY an ordinary AM broadcast receiver by retuning its local oscillator and RF stages slightly so that they tune higher and cover the cordless telephone frequencies. Just remove some capacitance and diddle the trimmer capacitors... Then if I'm right about its being AM, you should be in business. -John Sangster, W3IKG jhs at MITRE-Bedford