January 1957 Popular Electronics[Table of Contents]
People old and young enjoy waxing nostalgic about and learning some of the history of early electronics. Popular Electronics was published from October 1954 through April 1985. All copyrights are hereby acknowledged. See all articles from Popular Electronics.
According to a 2001 paper published by the National Institute of Standards and Technology (NIST, formerly National Bureau of Standards, NBS), "The end of the era of quartz frequency standards began in 1949 with the development at NBS of the world’s first atomic frequency standard based on an ammonia absorption line at 23.87 GHz." Further, "The Bureau supported work on both technologies for the next decade, but the rapid advances in the accuracy of atomic frequency standards could not be matched by quartz devices, and the work on quartz frequency standards was stopped in 1959." This article from a 1957 edition of Popular Electronics claims that the "master of all master-clocks" resided at the U.S. Naval Observatory at the time - not quite accurate from what my research indicates. Still, it is a good introduction to crystal growth and processing for use as timing devices.
Some Facts on Quartz Crystals
In the U. S. Naval Observatory at Arlington, Va., is the master of all master-clocks. The primary standard for all the timepieces of the nation, this clock is never fast or slow by more than a few seconds in one-hundred million. If you were to measure the distance between New York and Miami, Florida, with this kind of accuracy, the results would not be in error by more than the length of one standard cigarette!
Such fabulous micro-precision, matched only by the earth itself as it spins on its axis, is keyed to a tiny slab of crystalline quartz held under rigid temperature control in a special oven. Although it may soon be supplanted by an even more precise time standard (resonance of cesium atoms), the quartz crystal still remains the most important frequency-controlling device in existence today.
How They Are Cut.
Oscillating crystals are cut from so-called mother stones by high-speed carborundum wheels. Although most finished plates come from natural quartz prisms, modern techniques for growing mother stones in the laboratory have been perfected to the degree where our dependence upon nature soon will be unnecessary. Synthetic crystals are often superior to natural ones since the conditions during synthesis are held under rigid laboratory control.
Synthetic quartz mother stone was grown in a laboratory.
Bell Telephone labs.
A crystal exhibits piezoelectric activity. This means that it develops electric charges on its faces when compressed or otherwise distorted in shape. When connected in the grid circuit of a vacuum tube, sustained oscillations of one particular frequency are produced. Since the thickness of the crystal plate is an important factor in determining the frequency of oscillation, the cut blank is ground and polished until its mechanical resonant frequency arrives at the desired value; the thinner the crystal, the higher its resonant frequency.
Thin wafers are cut from large mother stones by multiple saw. Wafers must then be ground and polished.
James Knights Co.
At one time, only X-cut and Y-cut crystals were available. These terms indicate that the crystal slices are taken from the mother stone at right angles to the X and Y axes. The X-axis is a line joining two opposite corners of the hexagonal prism and the Y-axis is a, line joining the mid-points of the two opposite faces. (See Fig. 1.) Both of these cuts are subject to temperature effects, changing frequency over relatively wide ranges as the temperature varies.
Fig. 1. Axes of mother stone and orientation planes of X-cut and Y-cut crystal blanks. A hexagonal prism has six X and six Y axes.
Crystal blanks are being loaded into a planetary type of lapping machine prior to final polishing.
Between 1934 and 1949, vastly improved cuts were discovered by scientists connected with Bell Telephone Laboratories and RCA. Labeled AT, BT, V, CT, and DT, these crystals were ground from blanks oriented at complex angles with the axes of the mother stone. Finally, in 1940, the most stable quartz crystal ever devised was announced by W. P. Mason. Known as the GT-cut, this crystal shows no appreciable change in frequency from -25° C to +75° C; the cut remains virtually "on-frequency" over a range of 180° F!
Disassembled military type FT-243 (7620-kc.) pressure-sandwich type of crystal holder used extensively in communications equipment.
The kind of cut you might use depends upon the depth of your purse and its contents. X- and Y-cuts are quite inexpensive while a GT-cut for a specific frequency is still prohibitive in cost to experimenters of average means.
Characteristics. The most important single crystal parameter is its temperature coefficient. Since the frequency of a Y-cut crystal rises with temperature, it is said to have a positive coefficient; conversely, the coefficient of an X-cut crystal is negative because frequency drops with rising temperature.
For example, the temperature coefficient of a certain Y-cut crystal is given as +75 p/m/C°. This is read as "plus 75 parts per million per °C." It means that for every degree rise in temperature, the crystal frequency rises 75 cps for each megacycle of its basic frequency. Thus, if you assume that the temperature of this Y-cut crystal ground for 3 mc. at 0° C goes up 10° C, the change in frequency will be: 75 cps X 3 mc. X 10° = 2250 cps = .00225 mc. Adding this to the basic frequency gives 3.00225 mc. On the surface, this does not appear to be a serious deviation, but it is a change of 750 parts per million for only 10° of temperature variation. The GT-cut, on the other hand, varies only one part per million over a 100° C range (180° F).
Overtone Crystals. An overtone or harmonic quartz crystal is one that has been specially ground or otherwise treated by the manufacturer so that it vibrates in two or more parts rather than as a whole. Essentially, this process is very similar to overtone production in musical instruments where the sounding body vibrates in parts showing nodes and loops along its length (Fig. 2). If a crystal were to vibrate in two equal parts, the output would be exactly double the fundamental or "one-part" frequency. In practice, this is seldom the case, because overtone crystals do not "break up" into equal sections as they oscillate.
Fig. 2. Analogy showing relation of overtones to fundamental: (A) a string producing its fundamental tone by vibrating in one part; (B) first overtone or second harmonic of the string; (C) second overtone or third harmonic.
If, for example, an AT-cut crystal is treated to produce third harmonic output at 21 mc., it might be marked "7-mc. fundamental" This means that its harmonic frequency is approximately three times its fundamental; its output may differ from the true third harmonic by several megacycles. When such crystals are purchased for transmitter control, the buyer should know the harmonic output rather than the fundamental frequency.
For extra-precise control of frequency, crystal holders are often enclosed in thermostatically controlled ovens which maintain the frequency constant over extremely wide variations in ambient temperature. Such holders are very compact and weigh only a few ounces.
Four types of mountings. At left is a 110-kc. X-cut crystal in a military type HC-13/U holder which produces very little damping and some mechanical resonance in wires supporting the crystal. Next is a GT-cut in a special holder; the crystal is first plated with gold, then placed in evacuated glass holder, and is supported by eight wires soldered to plating. In cutaway view, a circular, silver-plated AT-cut in a glass holder is mounted inside temperature-controlled oven for greater stability. At right is a close-up view of an NT-cut in a miniature glass holder supported by four wires.
Posted April 3, 2013