July 1933 Radio-Craft[Table of Contents]
People old and young enjoy waxing nostalgic about and learning some of the history of early electronics. Radio-Craft was published from 1929 through 1953. All copyrights are hereby acknowledged. See all articles from Radio-Craft.
While reading through this article on copper-oxide rectifiers, I am once again reminded of how much we take for granted the conveniences of electrical test equipment on today's shop benches. The advent of FET-input multimeters was a huge step forward because the meter input impedance is so high that it has practically no impact on the circuit being measured. Prior to that, most simple meters drew their power from the circuit under test, thereby altering the true value of current or voltage being measured. Of course there were vacuum tube voltmeters (VTVM) with high input impedances, but few hobbyists or laymen could afford them. This piece reports on how the advent of a non-tube-based rectifier permitted alternating current (AC) measurements to be made by DC-driven d'Arsonval meter movements so as to not excessively load the circuit being measured. Rectox meters had the rectification components and calibration built in, and while still more expensive than circuit-driven meters, were much more affordable than other types. Today, of course, you can get a free DMM with more accuracy and greater functionality than any 1930s era multimeter at Harbor Freight just by handing the clerk a readily available coupon.
*Meter and Instrument Section, Westinghouse Electric and Mfg. Co.
The copper-oxide rectifier is used so extensively in modern test equipment, and is so little understood, that the editors are pleased to present this excellent discussion of the uses and limitations of copper-oxide rectifiers.
The practical measurement of alternating currents has, heretofore, been made by three types of instruments: the electrodynamometer, the repulsion or attraction iron vane type and thermo-couple-d'Arsonval type. The electrodynamometer type has a system of stationary and moving coils without iron in the magnetic circuits; the repulsion iron-vane types have a stationary coil, a movable vane and a stationary vane. Some types of this movement consist of a stationary coil and a single, movable attraction iron vane. In the thermo-couple-d'Arsonval type - a thermocouple is heated by the alternating current and the resulting thermo-emf is measured by a d'Arsonval instrument. Now we have a fourth, practical A. C. instrument known as the Rectox type. It consists of a copper-oxide rectifier and a d'Arsonval type of instrument. The alternating current is rectified and then measured by the ordinary direct-current d'Arsonval instrument.
The energy consumption, or the power required to operate the pointer; in the previous types of alternating-current instruments is much more than that required for direct-current instruments. This is because, in a direct-current instrument, the magnetic field is supplied from a strong permanent magnet, permitting a comparatively small current in the moving coil. The higher energy consumption of alternating current instruments has been an application handicap for a long time, especially, where the energy consumed by the instrument would seriously change the circuit conditions, particularly in radio measurements. The Rectox instrument has successfully solved this problem by embodying the high sensitivity features of the d'Arsonval type in an alternating-current instrument.
The rectifier used in the Rectox instrument is a product of Westinghouse engineering and research. It is specially designed for instrument use, the requirements of which are considerably different from the usual well known battery charging applications.
The rectifier units are plates of copper which have been oxidized on one side. Copper, when oxidized on its surface, has the peculiar property of rectification, allowing current to flow much more readily from the oxide to the copper, than from the copper to the oxide. The copper plates are assembled and held firmly in place, under constant pressure, by a sturdy clamp, and the entire unit is impregnated to seal it from moisture and corrosion. It is interesting to note that the size of these rectifier units is considerably smaller than the usual battery-charging unit. The relative area of the copper oxide plates greatly affects the performance of rectifier instruments; the proper area of plate has been determined after exhaustive research.
The assembly of copper plates is made to give full-wave rectification for all instrument applications. This is accomplished by assembly of the plates into four sections in reverse order and connecting to the instrument as shown in Fig. 1.
The Rectox instrument has certain characteristics which makes its application to measurement of alternating currents somewhat critical. For this reason its operating characteristics should be carefully considered in any application for measurement purposes.
This class of instruments differs from the usual alternating-current types in that the torque and deflection are proportional to the first power of the current. Therefore, it measures the average value and not the effective value of the alternating-current wave. The scale is, however, calibrated to read effective, or root mean square, values of a pure sine wave. Consequently, such instruments read correctly only on sine waves and have serious errors on other than sine wave forms. These errors can be compensated for in readings, provided the wave form is known from which a correction factor can be applied; or, the instrument may be calibrated on the wave form with which it is used.
The resistance of the Rectox is a function of the current flowing. This characteristic is shown in Fig. 2. When a rectifier-type milliammeter is connected in a circuit, it affects the circuit conditions because of its added resistance, like any other instrument; also the changes in the circuit depend upon the value of current passing. This disturbance of the normal circuit must be recognized if the milliammeter resistance is a large percentage of the total circuit resistance. If the circuit resistance is relatively high, then this change will result in negligible effects. The instrument always correctly indicates the actual current passing through the circuit; but the magnitude of the current may depend upon the non-linear value of the instrument resistance. (The actual resistance of the Rectox varying with current. - Editor)
The readings of Rectox instruments are quite free from frequency errors. The reading may be expected to decrease about 1/2 per cent per kilocycle up to 35,000 cycles, where different conditions occur. Because of capacity effects, this type of instrument is not recommended for radio-frequency measurements. It is, however, reasonably accurate throughout the audio frequency bands.
The effect of current upon the resistance of a rectifier unit has been previously discussed, but the copper-oxide unit also has the property of changing its resistance with temperature; and, furthermore, the amount of change due to temperature depends on the amount of current passing. Temperature resistance curves are shown in Fig. 3. We have, therefore, a very complex .relation between current, temperature, and resistance. As a result of these conditions, a great deal of skill is required in designing a Rectox instrument to prevent errors arising from temperature changes.
Tests show that the effective resistance of a copper-oxide rectifier decreases as the temperature increases. Therefore, if a rectifier instrument should be used as a low-range voltmeter, without suitable temperature compensation, the voltmeter might read as much as 20 or 25 percent high at a temperature of 40°C. It is for reasons of this kind that little success has been met in trying to adapt Rectox units to standard direct-current instruments. Much better results have been obtained by use of specifically designed combinations of instrument and rectifier, in which proper temperature compensation has been developed.
Like all devices of its kind, the efficiency of the copper-oxide rectifier is less than 100%; in other words, if 1 milliampere A.C. is passed through it, the resulting rectified current available for operating the indicating instrument is usually 8/10 of a milliampere, or less. The typical current efficiency curve for a rectifier instrument is shown in Fig. 4. Furthermore, the current-efficiency ratio (D.C. current output divided by A.C. current input) is affected to some extent by temperature and, also, by the absolute value of current flowing. Fig. 5 shows this characteristic. This again results in a complex situation involving temperature, current, and efficiency; and a simultaneous study of all of these variables is the only means by which errors from temperature variations can be minimized. For example, with certain values of current flowing, the efficiency of an uncompensated rectifier instrument may drop from 80% to 75% during a 40° C. temperature rise. This would result in lowering the calibration of the instrument 6% at the higher temperature if no steps were taken to secure temperature compensation.
The majority of the above discussed errors, characteristic of rectifier instruments, can be minimized by careful design and by taking advantage of the opposite effects of certain errors. However, it is important that the instrument be properly designed to operate with the copper-oxide rectifier. It is therefore, not advisable to try to apply a copper-oxide rectifier to an existing d'Arsonval instrument which has not been designed for this application unless changes are made to provide proper moving coil resistance, temperature compensation, and swamping resistance. If the errors are properly cared for in the design of a complete rectifier instrument, reasonable accuracy can be obtained.
The chief advantage of. a rectifier instrument is in its high sensitivity. By use of the rectifier principle, alternating-current voltmeters may be made with a very high resistance per volt. Standard voltmeters are available in ratings as low as 4 volts with a sensitivity of 1,000 ohms per volt, 1.5 volts with 2,000 ohms per volt, and even 0.5-volt with 5,000 ohms per volt. Below four volts, rectifier voltmeters should have a resistance of 2,000 ohms and, better still, 5,000 ohms per volt, in order to properly compensate for the errors discussed above. Milliammeters and microammeters of low ratings are also available.
Rectifier instruments are rapidly finding their place in the radio field for the measurement of such quantities as output of amplifiers and oscillators and power level indicators. The user of these instruments should bear their characteristics in mind, particularly their accuracy, when used under various conditions. Rectifier instruments are a valuable contribution to the science of radio and they are continually finding new uses in this rapidly advancing art. Possibly the further developments in research and engineering on these instruments will tend to minimize their present errors and make them still more useful.
Posted July 28, 2015