On Signal Strength Evaluation
October 1976 QST

It is always interesting to look back many years ago and see what people thought were the big issues of the day. According to this 1975 QST article, some amateur radio operators thought the decades old S-scale system for determining signal strength using a combination of objective and subjective measures was in need of modernization. Nine separate levels of signal strength was, in this author's estimation, too many to provide a meaningful report to the sender regarding how well his message was being received. Five was plenty. He had a good point since, as pointed out, can an operator really say whether a signal quality rated a score of S5 versus S6? History has been the judge of the matter, and now 44 years later the original nine levels has survived.

The S-meter and the S-scale - is it time for changing the rulebook? Could be!

By Jury A. Belevich*, UA1IG

A QSO without a signal report is a rare thing. And for good reason too - we all like to know how our signal sounds at the other end. Years ago, hams were just as concerned about their signals (if not more so), and a system, the RST (readability, signal strength, tone) system, was conceived to standardize signal reporting.

Is the RST System Outdated?

But through the years, the S-scale of the RST system has shown its age. The scale shown in Table 1, which spreads the estimation of signal strength over a very wide scale, may have been convenient when radio reception had only natural (cosmic and atmospheric) disturbances. Today, however, man-made and interstation (adjacent-channel) interference prevails. The S-scale is almost inaccurate as a result of this interference, because the difference between two adjacent reading points on the scale is so small. What is the difference, for instance, between an S4 and an S5 signal?

When estimating signal strength based on the current S-scale, differentiation between S2 (very weak) and S3 (weak) signals is difficult, if not impossible.

Better Signal Estimates with Modified S-System

We'll obtain more precision in signal estimation if we substitute instead a simplified S-system and the following S-scale. Some intermediate S-meanings have been removed from this scale so that it is simpler to estimate the signal strength more accurately.

It isn't long before most amateurs learn the great difference between estimating signal strength by ear and doing so with the help of a receiver's S-meter. This difference is caused by calibration variations and various S-meter amplifier tube characteristics.

For example, the S-meters on two receivers which have different sensitivity levels must be calibrated. It is recommended to estimate the S9 signal with an input signal strength of 100 microvolts when the S-units are spaced at 6-dB intervals. The calibration is made by the voltage applied to the input terminals of a receiver, as shown in Table 3.

With this method of calibration, S-meter scale graduation marks on a receiver with 1-microvolt sensitivity and a second receiver with 50 microvolts will proceed from S3 to S8 (Fig. 1A and B), because these signal strengths correspond to sensitivity of the receivers. With the same input signals (for instance, 100 microvolts) applied to the receivers, the S-meters will indicate the same signal strength (S9) because they are calibrated in the same way.

S-Meter Calibration Problems

There is a problem, though. When estimating signals by ear, an operator will determine S6-S7 in the first receiver and not more than S1-S2 in the second. This method of S-meter calibration allows for measurement of the signal level, but it doesn't reflect real conditions of communication, which depend not only on signal strength but also on the state of a receiver (and its quality too).

Often, a receiver has an S-meter amplifier tube with deteriorated characteristics. This causes understated S-meter readings, such as "RS-51" or "RS-52." This is full readability on a faint signal.

Another problem with measuring signal strength is that receiver manufacturers use various methods of S-meter calibration. For example:

1) The National Radio Co., Hallicrafters. The use of 50-m V input signal to determine the S9 point. Each S-unit equals 6 dB.

2) Swan Electronics Corp., Collins Radio Co. The standard for S-meters is 100 mV through a 6-dB pad for S9.

3) The R. L. Drake Co. The standard is 50 mV through a 50-dB pad for S9.

Suggested Solution

There is another method of S-meter calibration which this writer considers more correct. First, the zero level of the input signal is determined, the level which corresponds to the receiver noise level (it may also be called maximum or critical sensitivity level). In acoustics, this level is considered the "audibility threshold level." From acoustics, we know that normal speech exceeds the audibility threshold by 60-70 dB. Thus, we adjust a receiver's S-scale so that the maximum signal strength exceeds the receiver noise level by 60 dB. If we consider the maximum signal strength to correspond to S5 (Table 2), each S-unit will be equal to 12 dB. As receiver sensitivity contains the condition of necessary signal-over-noise excess to provide normal reception (usually 10-20 dB), this value conforms well to the concept of receiver sensitivity.

With this method of S-meter calibration, an aural estimation of a signal is the same as the S-meter reading. If the S-meter measures the value of the i-f voltage at the output of the receiver, the S-meter readings will remain accurate despite receiver deterioration (such as in the case of aging tubes). There will still be some dependence of the S-meter sensitivity on its amplifier tube if the S-meter measures the age voltage.

* a/box 20, Leningrad, M-244, 196244, USSR

Posted September 18, 2020