AN AMPLIFIER of some kind is an essential part of most instruments used for making fine measurements or controlling an apparatus or a process automatically. For example, the sensitivity required in an ordinary thermometer is developed by a hydraulic amplifier in the form of a capillary tube sealed to a glass bulb. Variations in temperature that cause almost imperceptible change in the volume of fluid in the bulb appear as pronounced excursions of fluid in the capillary. Other examples include the moiré verniers of the surveyor's transit, the train of gears in a dial micrometer, the jeweled levers in a mechanical seismograph and electronic amplifiers of various kinds in light meters, ion gauges, gas chromatographs and so on.

All good amplifiers are difficult to design. They are even more difficult to make at home, a fact that has discouraged many amateurs who otherwise might enjoy experimenting with sensitive instruments. A promising solution to this problem has recently appeared m the form of the small electronic devices known as operational amplifiers. They were developed to perform arithmetical operations in computers. Experimenters soon discovered that the amplifiers make handy building blocks for constructing sensitive instruments and control devices of all kinds.

The typical operational amplifier resembles a block of plastic about the size of a cube of sugar. On the back of the block are five or more short stubs of wire; they are the connecting terminals. In a five-terminal block one terminal is the input connection, another is the output, the third is a common terminal and the remaining pair are for connecting the unit to a source of direct current, such as a small dry battery.

The gain, or amplifying power, of the units is typically several million, far more than is required in practical applications. The gain can be reduced as desired by feeding some of the output energy back into the input terminal through a resistor. A second resistor is connected in series with the input terminal. The resistors establish the "closed loop" gain of the amplifier, which is exactly equal to the value of the feedback resistor divided by the value of the input resistor. For example, if the feedback is one million ohms and the input resistor is 1,000 ohms, the amplifier will increase the strength of the input signal exactly 1,000 times. The feedback scheme not only fixes the gain but also makes the amplifier relatively insensitive to variations of loading, temperature and other external influences.

The input signal can be either direct current or alternating current. The most sensitive operational amplifier can detect an input signal of a millionth of a billionth of an ampere and deliver an output of .02 ampere at a potential of 10 volts. This is much more energy than is required to operate an inexpensive meter or to drive a power amplifier.

Some types of operational amplifier have two input terminals but only one output terminal. Such amplifiers are known as differential amplifiers. They compare two independent input signals and amplify the difference between them. The devices are used, among other things, for determining an unknown voltage by comparing it with a known voltage. The known voltage is connected to one input terminal, the unknown voltage to the second input terminal. The output voltage divided by the gain of the amplifier equals the difference. By making the gain sufficiently large the experimenter can determine the value of the unknown voltage to any desired accuracy within the limits of the amplifier.

Quantities other than voltage can be similarly determined by equipping the amplifier with appropriate sensing devices. For example, if the input voltages are derived from a pair of strain gauges, the output will be proportionate to the difference between two forces [see "The Amateur Scientist"; SCIENTIFIC AMERICAN, January, 1968]. By substituting for the strain gauges a pair of photocells, thermocouples or glass electrodes, comparisons can be made in terms of such factors as light intensity, temperature and the acid-alkaline balance of chemical solutions.

Differential amplifiers are also used as the primary functional element in servomechanisms, or automatic-control devices. The principle of these mechanisms is illustrated by an apparatus recently designed and built by J. Barry Shackleford of East Point, Ga. He writes as follows:

"For some years I have been on the lookout for a graphic recorder I could afford, or plans for the construction of one. Commercial recorders of the quality I wanted cost $500 or more. The homemade varieties that have been described are not sufficiently sensitive or versatile for recording the data I wanted to collect. Several months ago I came across a description of an operational amplifier of the differential type and decided to try one as the central element in a mechanism for converting a varying voltage into the proportional displacement of a pen. The apparatus works so well that it may interest other experimenters.

"The mechanism consists of five essential parts: a differential amplifier, a current amplifier, a reversible direct-current motor, a potentiometer and a power supply. The differential amplifier compares a signal voltage with a reference voltage and amplifies the difference. This output is then increased sufficiently by the current amplifier to operate the motor. The motor drives a train of reduction gears that rotates the shaft of a potentiometer. The potentiometer develops a voltage that is proportional to the position of its contact arm, which is fixed to the shaft. This voltage is fed back to the positive input terminal of the differential amplifier as a reference potential for comparison with the signal voltage.

"When the signal voltage exactly equals the reference voltage, the output of the amplifier is zero. The motor receives no power and the contact arm of the potentiometer remains fixed. If the signal voltage now increases or decreases, the difference is amplified and appears across the terminals of the motor. The motor then rotates the contact arm of the potentiometer to the point where the reference voltage again exactly equals the signal voltage.

"In effect, the system converts variations of signal voltage into corresponding displacements of the contact arm. The displacements can be recorded in ink by attaching a pen to the outer end of a lever clamped to the shaft of the potentiometer. Some potentiometers are constructed in the form of a helix, so that the contact arm must make 10 or more revolutions to travel from one extreme to the other. By installing a pulley on the shaft of a potentiometer of this kind a belt can be used to displace a pen along a straight line. Other devices can be substituted for the pen, such as a valve, a rudder or a throttle.

"My differential amplifier operates from a direct-current source of 15 volts. It has six terminals: two (as a pair) for the power, one for the positive input, one for the negative input, an output terminal and a common terminal that is connected to the metal chassis on which the apparatus is mounted. Each input terminal is equipped with a pair of resistors for fixing the gain and providing negative feedback.

"In general operational amplifiers are adequately stabilized by a feedback current ranging from one microampere to 10 microamperes. I assumed a current of about four microamperes for the purpose of calculating the value of the feedback resistor. The resistance is equal to the voltage of the power supply divided by the assumed current: ohms. Since I happened to have several 3.3-megohm resistors, I used them because they were reasonably close to the calculated value.

"I decided arbitrarily to fix the gain of the amplifier at 100 for the initial experiment. The gain is equal to the value of the feedback resistor divided by the value of the resistor placed in series with the input. In this case the gain is 3.3 x 10/100 = 33,000 ohms. A resistor of this value was connected to each of the input terminals. Feedback resistors were then connected between the output terminal and the positive input terminal and between the negative input terminal and the chassis [see illustration at left]. The common terminal was connected directly to the chassis by a copper lead. With the exception of the two input connections and the battery connection, this completed the wiring of the differential amplifier.

"When the signal voltage is lower than the reference voltage, the amplifier develops a negative voltage at the output terminal. On the other hand, positive voltage appears at the output terminal when the signal voltage is higher than the reference voltage. The amplifier delivers a maximum output current of .02 ampere. The motor I used requires a current of one ampere. To develop the necessary additional current I inserted a current amplifier between the differential amplifier and the motor. Current amplifiers are available commercially but can be made so easily and inexpensively that I assembled my own.

"The current amplifier uses two power transistors; they are identical except that one amplifies negative input voltage and the other one amplifies positive input voltage. Any transistors of this general type can be used provided that they are designed for an output current of at least one ampere. Each transistor must be mounted on a heat sink-a metal plate with cooling fins for radiating heat that develops internally when the device conducts current.

"The general wiring scheme of the current amplifier is depicted by the accompanying schematic diagram [right] The output voltage of the differential amplifier is applied to the input circuit of the power transistors through diodes connected so that positive voltage causes one transistor to conduct and negative voltage causes the other one to conduct. The polarity of the-. output is positive when the first transistor conducts. The motor then rotates in one direction. When the other transistor conducts, the output is negative and the: motor reverses.

"The output lead of each transistor contains a 10-ohm resistor that limits the current and protects the device. Input voltage for triggering the transistors is developed across a pair of 15,000-ohm resistors connected in series with the diodes and the differential amplifier. Capacitors connected between the power-supply terminals and the chassis act as filters for suppressing spurious alternating currents.

"The reference voltage is derived from the power supply, which is connected to the fixed terminals of the potentiometer. The reference voltage is applied to one input terminal of the amplifier by a lead from the sliding contact of the potentiometer. The reference voltage should be somewhat larger than the expected range of the signal's voltage. For example, a reference potential of one volt is appropriate for a signal that varies from approximately .3 to .7 volt. The response of the potentiometer must be linear, that is, it must develop equal increments of reference voltage as the sliding contact advances through equal increments of arc.

"The apparatus can be operated by three dry batteries. Two of them, of about 15-volt capacity, are for supplying the amplifiers, and one 1.5-volt battery is for energizing the potentiometer. When batteries are used, the positive terminal of one 15-volt battery is connected to the negative terminal of the second 15-volt battery and to the chassis. The remaining battery terminals are connected to the apparatus.

"For reasons of economy I substituted a pair of power supplies for the 15-volt batteries. The units were regulated automatically to deliver exactly 15 volts apiece. Each unit consists of a small, inexpensive transformer that operates on house current to supply alternating current at 25 volts.

"The output of the transformer is converted to direct current by a rectifier of the bridge type that requires four diodes of the kind that will conduct an ampere and withstand a potential of 50 volts. The output of the rectifier is filtered by a pair of 500-microfarad electrolytic capacitors. Connected across the output of the rectifier is a voltage divider consisting of a 1,200-ohm resistor in series with a Zener diode.

"The Zener diode is in effect a variable resistor that automatically maintains fixed voltage in a circuit. Variations of potential cause the conductivity of the Zener diode to change in the direction required for maintaining constant voltage across its terminals. In my application the fixed voltage of the Zener diode controls the conductivity of a transistor in one lead of the power supply. Variations of the rectified voltage are instantly compensated by opposing variations of resistance in the transistor and vice versa, with the result that a constant potential appears across the output of the power supply. The power supplies are connected to the apparatus just as though they were dry batteries.

"The fully assembled apparatus performed much as I had hoped. When I applied a signal of .5 volt to the input terminal, the potentiometer promptly generated a corresponding reference voltage and thereafter followed almost every variation of the signal. I say 'almost' because the device is not quite perfect. A certain minimum signal is necessary to start the motor. When the motor is at rest, the input voltage must increase or decrease slightly before the amplifier develops sufficient power to start the motor.

"I call this the 'dead zone.' It can be narrowed somewhat by increasing the gain. If the gain is increased above a~ certain value, however, the motor tends to overshoot: to start abruptly and coast beyond the point where the reference voltage equals the signal voltage. It then reverses, overshoots in the opposite direction and continues to oscillate for a time around the desired point before coming to rest. Insufficient gain increases the width of the dead zone.

"Optimum gain can be established most easily by providing the differential amplifier with a variable gain control. This step can be accomplished by substituting for the feedback and-series resistors a dual potentiometer of about 3.5 megohms. The sliding contacts of the potentiometer are connected to the input terminals of the differential amplifier. One set of fixed terminals serves as the inputs. The fixed terminals of the other set are connected respectively to the output and to the chassis. With this arrangement the gain is adjusted to the point where the motor starts to oscillate and then is decreased slightly. The width of the dead zone could also be reduced by using a more sensitive motor.

"My motor was salvaged from a portable tape recorder that operated on dry cells. Several extraordinarily sensitive direct-current motors that are equipped with permanent-magnet fields have been made to operate from solar cells by the International Rectifier Company. Doubtless a motor of this type would improve the performance of the apparatus.

"The motor must be coupled mechanically to the reference-voltage potentiometer through a set of reduction gears. I used a surplus unit that works fine. Any gears that are relatively free of friction can be substituted. The gear ratio of my unit enables the motor to rotate the potentiometer to either of its extremes in about a second, a response time that is adequate for recording seismic waves, changes m the brightness of stars and much other information of interest to amateurs. The operational amplifier I used is Type 118-A. It is made by Analog Devices, Inc., 221 Fifth Street, Cambridge, Mass. 02142. It costs $11."

RECENTLY the editor of this department came across two other new and relatively inexpensive products of possible interest to amateurs. Both were developed for determining the acid-alkaline balance (pH) of materials by means of dyes that change color. The basic technique is not new. The interest of the new devices lies in the kinds of materials that can now be tested and in the accuracy of the measurements.

One of the devices measures the pH of the surface of such solids as paper, plastic, metal, textiles and minerals. This has not been possible heretofore. Often experimenters apply a printed circuit or some other metallic film to a substrate such as a sheet of plastic only to discover, when the apparatus fails some weeks later, that the surface was either acidic by nature or that it had been contaminated by alkali. Similarly, filter paper of the kind used for making chemical separations is supposed to be neutral but often turns out to have been contaminated by acid fumes.

The device for measuring the pH of solids consists of a set of four crayons] contained in refillable mechanical pencils. To measure the pH of a surface you draw a mark on the material, dab it with a wad of cotton moistened with distilled water, wait 15 seconds and compare the hue of the mark with a scale of color that comes with the pencil. The crayons span the range of pH from 1 to 12.

The second product of interest is a new test paper containing a dye that is remarkably sensitive to liquids of low ion concentration. Dyes that are customarily used tend to react with the solution being tested and, if the concentration of ions is low, to alter pH. For this reason, conventional test papers can result in an error of as much as a full pH unit. The new material reduces the error to less than a quarter of a unit. It consists of a roll of paper tape in a dispenser, a test tube and a color chart. The test is made by immersing a strip of the paper in a container of specimen solution and, after one minute, comparing the color of the strip with the color chart. The pH range of the dye is from 3 to 10. Both test kits are products of Micro Essential Laboratory Inc., 4224 Avenue H, Brooklyn, N.Y. 11210.

Bibliography

HANDBOOK OF OPERATIONAL APPLICATIONS. Burr-Brown Research Corporation, Tucson, Arizona.

The Society for Amateur Scientists (SAS) is a nonprofit research and educational organization dedicated to helping people enrich their lives by following their passion to take part in scientific adventures of all kinds.