1996 - 2016
BSEE - KB3UON
RF Cafe began life in 1996 as "RF Tools" in an AOL screen name web space totaling 2 MB. Its primary purpose was to provide me with ready access to commonly needed formulas and reference material while performing my work as an RF system and circuit design engineer. The Internet was still largely an unknown entity at the time and not much was available in the form of WYSIWYG ...
All trademarks, copyrights, patents, and other rights of ownership to images and text used on the RF Cafe website are hereby acknowledged.
My Hobby Website:
October 1947 QST
'QRM' is the Q-code in Ham-speak for unwelcomed manmade inband electrical interference. Interference is not just random signals like noise from motor brush arcing, intermittent electric distribution system connections or inter-conductor arcing, etc.. An improperly tuned or ineffectively filtered radio transmission, or EM energy leaking from a poorly shielded electronic device is also QRM. I distinguish such noise as unwelcomed because what might be considered as noise by one person could be a desired signal by another. 'QRN' stands for electrical noise generated in nature such as lightning bolts, solar storms, or even, as discovered by Drs. Arno Penzias and Robert Wilson, the 160 GHz Cosmic Microwave Background (CMB) radiation that emanates from all regions of the sky. A mnemonic for remembering which Q-code is which is the trailing 'M' for manmade and 'N' for natural.
The interference elimination scheme described here works by using a local oscillator to generating a second, lower intermediate frequency (IF) that then is fed through a narrow filter on either the upper or lower sideband, thus eliminating nearby interferers. It is fundamentally a dual conversion receiver, which is still used when suppressing strong multitone signals required - particularly where purely analog signal are used (hence, not much opportunity for post-detection signal processing).
Selectable Single-Sideband Reception Up-to-Date
By J. L. A. McLaughlin
Here is a simplified and improved version of the receiving system first introduced in QST just before the war. A thorough trial in wartime radio intelligence work proved the worth of the system - a system that can go a long way toward eliminating QRM in both phone and c.w. reception.
The need for improved means of receiving signals through heterodyne beat-note interference has in the last few years become increasingly apparent.
During the war the writer designed and built for the Federal Communications Commission and the Office of Strategic Services a receiving system that enabled them to copy phone and c.w. transmissions through terrific heterodyne QRM that made reception hopelessly impossible on the best. conventional receivers.
The FCC first employed this communications aid as far back as the summer of 1941. The June, 1941, issue of QST contained an article by this author describing this communications development.1 Mr. George Sterling, then chief of the Radio Intelligence Division of the FCC, was quick to recognize the importance of this invention2 to the highly specialized work in which the Commission was engaged. The Commission immediately purchased the original development model and subsequently ordered units for all primary monitoring stations throughout the country. Because of Mr. Sterling's foresightedness, when war came one Government agency, at least, was capable of carrying on radio intelligence work in the face of malicious or accidental interference. When the communications division of the OSS was set up, shortly after the start of the war, it, too, promptly ordered similar equipment for its services.
The first war model supplied the OSS and the FCC was similar to the early models used by the FCC. Later, a second war model was designed for the OSS; it was a decided improvement over earlier models both in performance and design. It was more compact, for one thing, and it was self-contained and could be connected to any of the standard communications receivers in use by the OSS, without modification or circuit changes in the attached receiver. Because this later model lends itself more to present-day amateur requirements, this article will be devoted to an explanation of its performance characteristics in the presence of strong heterodyne interference.
How this new heterodyne-eliminating receiver operates will perhaps be made clearer if we take up first the causes of beat-note interference and the inherent weakness of todays communications receivers in the presence of such interference.
The single heterodyne audio beat note, the product of one off-frequency carrier boating with the carrier of the desired signal, is well understood, but the audio beats produced by multiple off-frequency carriers are not clear to many.
Fig. 1 will help to form a picture of just what takes place after rectification of two or more carriers. Fig. 1-C indicates that when four carriers are present six principal audio beat notes are produced by rectification.
The removal of one heterodyne beat note can be achieved either before or after rectification by some form of phasing device; that is, some scheme capable of putting a variable rejection notch in the response curve of either the i.f. or a.f. amplifiers. Schemes such as these have been mentioned in the pages of QST by this and other authors. The rejection of a single interfering carrier can be demonstrated quite beautifully in the laboratory, but under normal communications operations, when complex heterodynes are present, these systems fail to generate any great enthusiasm in the operator. The reason for this coolness can be found in an inherent weakness in all such devices - that is, in the presence of heterodyne interference the beat-note tone seldom will give any clue as to whether or not it is being produced by only two carriers, or by more than two. If there are more than two carriers present this sort of rejector falls down. Instead of being an aid the adjustable rejection becomes a nuisance, and distracts the operator's attention from the real job at hand - i.e., the message being received - and forces his attention on the beat notes.
It is obvious that to be useful under present-day crowded band conditions any practical system of heterodyne elimination must first of all be rapid in operation, suppressing all the interference that it is capable of suppressing under the particular receiving conditions in a minimum of operating time. It must not introduce any new operating techniques alien to the normal training of the operator - rather it must permit the operator to concentrate on the' signal being received, not on the interference.
The system developed by this author (Fig. 2), which is the subject of this article, satisfies these conditions. It is fast and effective, being semiautomatic in eliminating multiple-heterodyne QRM both on 'phone mid c.w.
The receiver is fundamentally a triple-detector superheterodyne. The desired signal in the first i.f. system (455 kc.) is converted to a new intermediate frequency of 50 kc. This 50-kc. i.f. system differs from the conventional in that the response curve is unsymmetrical (Fig. 3). All frequencies below the carrier (50 kc.) are greatly attenuated, giving the amplifier the characteristics of a high-pass filter.
On 'phone reception this unsymmetrical selectivity of the 50-kc. i.f, system permits single-sideband reception. Since both sidebands contain identical intelligence, we can sacrifice the one containing the undesired signal without reduction of intelligibility or naturalness.
The manner in which the desired single sideband is selected is as follows: Two crystal-controlled oscillators are used, one ("A") on 405 kc. and the other ("B") on 505 kc. Either will convert the 455-kc. carrier to 50 kc. Although the desired carrier remains the same in both cases, all other frequencies converted will be transposed when switching from oscillator "A" to oscillator "B." "A" converts the 455-kc. signal to 50 kc. and all the side frequencies in the same numerical order, hence the upper single-sideband frequencies are selected in this case. Oscillator "B" converts the 455-kc. signal to 50 kc. and inverts the numerical order of the sideband frequencies, hence the lower sideband frequencies are selected in this case.
Assuming that an undesired carrier happens to be 456 kc., "A" will convert this "side" frequency of 456 kc. to 51 kc., and oscillator "B" will convert the same frequency to 49 kc. In other words, we have here a system in which we can switch undesired carriers from a frequency on one side of the desired carrier to a new frequency on the other side. Since the 50-kc. i.f. is of the high-pass single-sideband type, this switch permits placing the undesired carrier either in or out of the passband frequencies. In the case of the 456-kc. interference, oscillator "B" would be selected to eliminate the 1000-cycle beat note; "B" converts the signal to 49 kc., which frequency is attenuated 50 db. in the 50-kc. i.f. filter. If "A" had been used instead, the undesired signal would have been converted to 51 kc., resulting in no attenuation at all.
C. W. Reception
The selectable single-sideband system of heterodyne elimination is an obvious improvement in the reception of 'phone signals. At first glance its value in c.w. operation may not be so apparent. The improvements, though not obvious, are nevertheless present. The unsymmetrical filter (50-kc. i.f.) cuts off very sharply at the edge of the signal carrier's frequency; it is similar to a crystal filter with the rejection notch set about 1000 cycles below resonance. It differs from the crystal curve, however, in that it cuts off a wide band of frequencies rather than putting a notch at one particular frequency in the resonance curve. By means of the sideband selector switch we can flip an undesired carrier to the low-frequency side of the unsymmetrical filter. It should be obvious that throwing a switch that removes a whole band of frequencies is faster and easier to do than adjusting a critical phasing control, as is the practice in crystal-filter operation.
The second point in favor of this system over the crystal filter is that the objectionable" ping" of the high-Q crystal circuit is absent. A final improvement in the reception of C.W. signals is achieved by use of a sharply-tuned 1000-cycle filter in the audio circuit. This filter, together with the unsymmetrical response-curve switching system, makes for very easy c.w. operation even in the presence of tough QRM. In c. w. work the b.f.o. is left fixed at the correct frequency to produce a 1000-cycle beat note with the desired signal. The operator merely tunes for maximum signal strength.
Tuning the Carrier
A prime requisite of single-sideband 'phone operation is placing the desired carrier correctly in the bandpass filter of the second i.f. In the model described earlier a visual system of tuning was employed, using a tuning meter connected to the output of a sharply-tuned 50-kc. amplifier.1
In the later system this extra equipment has been eliminated and an accurate aural system substituted. The center position of the sideband selector switch is marked "carrier." In this position oscillators "A" and "B" are both operating, and the correct tuning is indicated aurally when the signal is tuned to zero beat with itself. (The two i.f. signals produced by the beats between the desired carrier and the two oscillators move in opposite directions as the receiver is tuned.) Further help in aural carrier positioning is achieved by narrowing the bandwidth of the high-pass filter in the "carrier" position of the switch. This bandwidth is made only a few hundred cycles wide and peaked sharply at 50 kc. When the sideband control switch is flipped either to the upper or lower sideband the original bandwidth of the high-pass filter is restored and one oscillator is disconnected. This improved aural tuning system permits normal tuning by ear of a 'phone signal in the presence of extreme interference.
For c.w. reception as well as 'phone the FCC and the OSS found this system far superior to the conventional communications receivers. These units made it possible to copy signals through heterodyne interference that otherwise would have made them unintelligible.
1 McLaughlin, "The Selectable Single-Sideband Receiving System," QST, June, 1941.
2 U. S. Patent No. 2,364,863.
Posted August 16, 2016