April 1939 QST
Antenna radiation (beam) patterns published by manufacturers are obtained under ideal - or close to ideal - conditions with a carefully prepared and calibrated open air test site (OATS) or an enclosed anechoic chamber. Multipath, imperfect earth ground, obstacles both manmade and natural, misshapen elements, poor VSWR, antenna orientation (in both azimuth and elevation) are among the many factors which produce real-world operational results that do not jive with a manufacturer's datasheet. Without employing some far field 3-dimensional field strength scheme (see Drone-Based Field Measurement System™), there is no way to obtain a complete picture of how your antenna performs in all directions. This article presents a practical procedure for making measurements that will at least collect useful data for specific orientations.
Converting Meter Beadings to Decibels for Signal-Strength Comparisons
By S. Gordon Taylor,* W2JCR
Many hams recognize the possibilities in this latter application but fail to take advantage of it because they (1) are unfamiliar with the procedure, (2) are reluctant to undertake the calculations ordinarily involved, or (3) lack data from receiver manufacturers which permits variations in S readings to be interpreted in terms of decibels, power ratios or voltage ratios.
It is the purpose of this article to provide data on a number of standard receivers, describe simple methods of checking characteristics and plotting patterns of rotary beams, and provide a chart which reduces calculations to a matter of simple arithmetic - with not an awful lot of that.
If you have a rotary beam and ask different stations for checks, the results will normally be of only the most general value because for the most part these reports will be given in terms of the number of S points difference between your minimum and maximum signal. A three-S difference on one standard receiver may, in terms of power ratios, equal a difference of five S's on another receiver, four S's on a third, and so on. For this reason the S reports from different receivers are not directly comparable, and it is impossible to arrive at any sort of average suitable for use in plotting antenna characteristics.
Nor can this problem be solved by arbitrarily assuming some standard db value per S point, a practice which is quite common. Measurements show that the "db per S" may average anywhere from 3 to 6 on different standard receivers. This is another way of saying that with one receiver model S9 may be around 50 db "up" from S1 while in another model it may be only 25 db up. What is more, the db difference between S3 and S4, for instance, may be quite different from that between S8 and S9 on the same receiver and S meter.
What is needed is knowledge of the actual db calibration of the S meters for the different receiver models, and this data will be found in Fig. 1 for nine standard receivers. The data on which this chart is based was obtained from the individual manufacturers, for this purpose (or from their literature in some cases), and most of it appears in print here for the first time.
The utility of this chart is obvious. It serves to provide a useful calibration for owners of any of these receivers. Of equal importance, it enables the ham to interpret, in terms of db, the comparative reports received from others who are using any of these receivers.
With rotary beams sprouting like mushrooms, the receiver S-meter takes on a new importance in furnishing a means for giving information on radiation patterns - provided its readings can be reduced to some standard. The important thing, of course, is the relative signal strength, easily expressible in terms of decibels. Since no two receiver S-point calibrations are alike, the information in this article is particularly timely and useful, and gives the beam owner a means of correlating signal reports.
Perhaps your receiver has a meter which you installed yourself and on which you scaled off your version of the S scale. In that case you can obtain the decibel calibration by comparing your readings on given signals with those of a friend who owns one of the receivers of Fig. 1. This comparison must of course be made with the two receivers in the same location and switching the same antenna. It will be most simple if you calibrate your meter to correspond with the S scale of the other; then its decibel calibration as given in Fig. 1 will apply to yours also.
It is perhaps well to point out that the varying heights of the columns in Fig. 1 have nothing to do with the relative sensitivity of the different receivers. In each case the column height simply represents the relative values which each receiver manufacturer chooses to employ in designing and calibrating his S meter circuit, and it is obvious that the different scales are not in agreement. The fact that some calibrations start at S° and others at S1 is of no importance; this again simply represents the manufacturer's choice of zero db level but does not in any way alter the utility of the db data of this chart.
When the owner of a rotary beam receives reports on his front-to-back ratio in terms of decibels these reports from different stations can be directly compared with a far greater degree of accuracy than would be the case were the reports simply given in S points. For example, suppose two receiving stations gave him reports. One, equipped with a "Sky Challenger" receiver reports readings of S9 for the front and S3 for the back. The other, using an "NC-100" reports S9 for the front and S6 for the back. On the basis of straight reports (without the data of Fig. 1) it would appear that the signal variation as observed at the first station is twice as great as at the second. Convert both to decibels, however, and the reports are nearly the same - 19 1/2 db and 18 db respectively.
A Translation Scale
Fig. 2 provides another tool of great value in checking beams and plotting radiation patterns. Here the decibel scale again appears, and related directly to it are power- and voltage-ratio scales. In the last column is the S scale of the Skyrider SX-17, which happens to be the receiver used at W2JCR. Those having other receivers should substitute the appropriate S scale.
To illustrate the use of the chart, assume that a rotary beam is being checked. "On the nose" a reading of S8 is obtained and off the back end the reading is S5. The db equivalents taken from column 2 are 36 and 21 and the difference is 15 db. Now referring to column 1 it is seen that 15 db represents power ratio of slightly over 30, and this is the front-to-back ratio of this particular beam. Thus a specific and decidedly useful report can be given to the station being checked.
The power ratio could be arrived at immediately without the intermediate conversion of the S readings to db, but if this is done we get a ratio of something like 3600 to 120, which is rather unwieldy. To reduce it to the simpler terms involves some mental gymnastics which are much more arduous than the S-db-power conversion scheme. Although beams are evaluated by hams almost entirely in terms of either db or power ratios, there are occasions when the voltage ratios are desired and for that reason they are included in the chart.
Connecting his rotary beam to his receiver, the owner can check its characteristics by tuning in some other station and noting the meter readings as the beam is rotated. The conversion to terms of power ratio is then made as described above.
Obviously the only additional work necessary for plotting the radiation pattern of a beam is to take a number of readings as the antenna is rotated, instead of just the two needed to check front-to-back ratio. There are several precautions to be taken in such checks, however, and the procedure followed at W2JCR may prove helpful.
The first thing is to determine the antenna position which puts the strongest signal into the receiver. If this is above S9 the antenna coupling to the receiver should be reduced so that the meter reads S9 or slightly lower, thus keeping within the range of the db calibration. Having found this position for the beam, and with the receiver antenna coupling suitably adjusted, the exact course of procedure is then agreed on between the two stations: The beam to be rotated in steps of not more than 22 1/2 degrees and to be stopped in each of these positions for at least 30 seconds; at each stop the position of the beam to be announced, then the carrier left unmodulated for the balance of the period.
At the receiver end the selectivity should be high to reduce the possibility of QRM, because an interfering signal will make the check valueless. Leaving the antenna coupling at the adjustment mentioned above, at each position of the beam make a note of the announced direction, then during the unmodulated period make sure the signal is exactly in tune and note the resulting reading. This retuning is important because changes in signal strength may otherwise tend to throw the receiver oscillator slightly out of tune, making readings unreliable. If modulation were present, that too might confuse the readings by "wobbulating" the meter.1 The meter readings are noted in fractional S points; usually it is possible to estimate to one-tenth of a point.
Table I is the work-sheet of an actual check made by W2JCR and will be briefly described to illustrate the orderly method of recording the measurements and compiling the desired data.
The directions as announced were entered in column 1 and the S readings in column 2. The transmitting station was then asked to stand by for a few minutes while the desired information was worked out in columns 3, 4 and 5.
In column 3 the db values for the various readings are noted. These are taken from the second column of Fig. 2. We note that the lowest value here is 15 db, and inasmuch as we are interested only in the relative values shown by this particular beam we adopt this as a new zero level and deduct 15 from each of the values in column 3, entering the result in column 4. From these figures we see that the maximum signal is 30 db above the minimum; that the "head-on" signal is 13 db above the back (southwest) signal, etc. Incidentally, we note that this beam is apparently not functioning "according to Hoyle" because it should have minimum radiation off the ends, which would be the SE and NW positions.
In this particular report the power ratios are entered in column 5. Had the owner of the beam been interested in voltage ratios these would have been shown instead. In either case the data are obtained directly from Fig. 2. If a power unit value of 1 is given to the minimum signal position, then the maximum signal, which is 30 db up, would represent 1000 times this power, etc.
With these figures completed a detailed report was given to the owner of the beam by reading off to him the figures of columns 4 and 5. Later, as a matter of interest, the pattern of his beam was drawn up as shown in Fig. 3, using polar graph paper.2 Lacking this it could have been drawn on plain paper on which a dot is first placed to represent the transmitter and lines drawn radiating from this at angles of 22 1/2 degrees - like the cuts of a pie that has been divided into 16 pieces.
The scale selected for use depends on the ratios involved. In this case the maximum is 1000 to 1, so the scale used was 200 per inch, making the front lobe 5 inches long. The other values were laid out on each of the lines corresponding to the beam positions and the resulting points were joined together with a line and this made up the pattern of the beam.
Had more readings been taken, perhaps at every 10 degrees, this pattern would have been less angular in appearance and more truly representative, because obviously the actual radiation pattern is not likely to have sharp points such as those shown at the front and back. Even in drawing the present pattern it would have been entirely legitimate to round off the corners; in fact this is usual practice although it is easier to draw as shown in Fig. 3.
Attention has been centered, in this discussion, on the use of the S meter and the data of Figs. 1 and 2 as aids in checking beams. But they have other applications as well. Power gain or loss for different degrees of antenna coupling at the transmitter, and different adjustments of the transmitter or transmitting antenna can be checked in the same way providing the" before and after" readings are made under similar conditions of receiver adjustment, line voltage, etc. The effectiveness of any two fixed transmitting antennas can likewise be determined, especially if they are so arranged that the transmitter can be switched from one to the other quickly. Such comparative checks obtained from several receivers located in different directions from the transmitter will provide helpful information, but the radiation pattern of a fixed directional beam can be determined in this manner only if the antenna used for comparison is non-directional or of known directional characteristics.
In conclusion, it is well to emphasize the fact that measurements such as those described in this article are not perfect. The human element plays an important part, as in accurately reading the meter, determining the exact angle between different positions of a beam, etc. A bad tube in the receiver may alter the meter calibration, but fortunately many of the ailments to which receivers are at times heir, while they may change the reading of the meter for a given signal voltage input, do not materially alter the relationship within the scale itself and therefore do not change the db calibrations given in Fig. 1. In any event, the system outlined represents the most accurate, generally available method of checking and is far superior to the old system of "three S's down," "two S's up" now in common use.
In plotting radiation patterns it should be borne in mind that two patterns plotted on the same beam but by different receivers in different directions will not necessarily be the same. There are numerous factors, such as reflection, absorption, and refraction, which may differ in different directions and different locations. But by obtaining checks from several different stations it is usually possible to strike some sort of average which will not only indicate the characteristics of the beam but may indicate specific faults such as a nearby structure which is affecting its operation.
* The NC-100 utilizes an electric eye in conjunction with the r.f. gain control, which has a 0-10 scale. With a signal tuned in the gain control is backed off until the signal barely registers on the eye, then the S reading is taken from a curve plotted in terms of gain-control settings. The combination is therefore the equivalent of a meter.
** The Super-Pro meter is not calibrated in S-points but instead has a 0-5-milliampere scale. With the antenna disconnected and no signal tuned in this will read somewhere around 4. As signals are tuned in the reading drops and it is the amount of this drop from the normal no-signal level that appears in the figure. Thus if the normal level is 4.1 and a signal drops it to 3.6, then that signal is 15 db above the "zero" level, etc.
1 Obviously, when making such a check with a distant station, it is also necessary to take the effects of fading into account and, if necessary t make the measurements over a sufficiently long period so that fading averages out. The measurements in such a case should be attempted only when conditions are relatively stable. - Editor.
2 Polar Coordinate paper No. 358-31. Keuffel and Esser, 127 Fulton St., N. Y. C.
Posted March 31, 2016