US2159796A - Quartz piezoelectric element - Google Patents

Description

May 23, 1939. H. w. N. HAWK QUARTZ PIEZOELECTRIC ELEMENT Filed Nov. 30, 1937 Lltyorneg Patented May 23, 1939 UNITED STATES PATENT OFFICE Henry W. N. Hawk,

Woodlynne, N. J., assignor to Radio Corporation of America, a corporation of Delaware Application November 30, 1937, Serial No. 177,291

8 Claims.

My invention relates to the piezo-electric art, particularly to so-called thickness-mode quartz piezo-electric elements and constitutes an improvement of the invention disclosed in my copending application Serial No. 128,054, filed February 27, 1937, now Patent 2,119,848, June 7, 1938.

In my above identified disclosure, it is shown that if a quartz blank dimensioned to respond to a desired thickness-mode fundamental frequency is provided with a convex electrode surface, of certain curvature, the finished element will be rendered capable of vibrating, selectively, at one or more harmonically related frequencies in addition to the said fundamental frequency. It is thus possible to achieve a piezo-electric element which will actually vibrate or oscillate at frequencies therebefore impossible of practical achievement.

The invention described in my earlier disclosure is shown to be applicable to crystal blanks of various cuts, including those out in the manner described in copending application of Bokovoy and Baldwin, Serial No. 69,496, filed March 18, 1936, (see also the corresponding British Patent 457,342) to exhibit a desired low or zero temperature coefficient of frequency. A blank possessing a zero temperature-frequency-characteristic and provided with a convex electrode surface, in accordance with my earlier invention, will exhibit that characteristic only when vibrated at its fundamental frequency.

Accordingly, the principal object of my present invention is to provide a plural frequency thickness-mode quartz piezo-electric crystal element which shall exhibit a zero, or substantially zero, temperature coefficient of frequency when vibrated at a frequency higher than that of its fundamental thickness-mode frequency.

More specifically, my present invention contemplates, and its practice provides, a piezo-electric element which will exhibit a zero or other desired low temperature coefficient of frequency when vibrated at a frequency which is three times that dictated by the thickness dimension of said element. Thus, when my present invention is applied to a blank dimensioned to vibrate at a fundamental thickness-mode frequency of, say, 5 megacycles per second, the finished element will exhibit the desired temperature coefficient of frequency only when vibrated at 15 megacycles per second.

The fact that the finished element when vibrated at its fundamental frequency will exhibit a temperature-frequency characteristic other than that exhibited at the third harmonic is, generally, of little importance for the reason that the crystal elements of my present invention find maximum usefulness in oscillator (or resonator) circuits designed to respond to frequencies beyond 5 the range of vibration of crystals now commercially available. Further, an element cut to exhibit a zero temperature coefiicient of frequency at its third harmonic will exhibit a variation with temperature of but a few cycles, per million, per 10 degree centigrade, when vibrated at its fundamental (or higher than the third harmonic) frequency, so that it may be usefully employed in circuits designed to respond, selectively, to any of the frequencies at which it is capable of being vibrated.

The method employed in preparing crystals in accordance with my invention, and further objects and advantages attained in carrying out the invention, will be more readily understood from the following description, taken in connection with the accompanying drawing, in which Figure 1 is a cross-sectional view of a quartz piezo-electric element having a convex electrode face, the degree of curvature being exaggerated.

Figure 2 is a partly diagrammatic perspective view illustrative of the manner of manufacturing the crystal element of Fig. 1,

Figure 3 is a top plan view of a quartz mother crystal with certain of its axes marked as an aid 30 to a clear understanding of the system of orientation followed in carrying the invention into effect,

Figure 4 shows the mother crystal of Fig. 3 in outline and in perspective, and having a section 5 cut and divided to provide a rough bar having top and bottom surfaces lying in planes which are normal to the Z (optic) axis and its long edges parallel to a W (reference) axis which is 25% removed from an X (electric) axis,

Figure 5 is an elevational view showing the position of a blank cut from the bar of Fig. 4, in accordance with the invention, prior to being finished, and

Figure 6 shows the blank of Fig. 5 removed and trimmed.

Fig. 7 shows a modification.

Thickness-mode crystal elements of the prior art are in the form of blanks (which may be square, round, polygonal, rectangular or other 0 shape) having duplicate, parallel, flat electrode surfaces. The objects of the present invention are achieved by starting with a blank of a certain later specified orientation and so finishing the blank that it will have at least one convex or convex-spherical electrode surface. Where the element is of bio-convex spherical contour, the radius of curvature of each electrode surface must be substantially exactly the same. Because of the difficulties incident to the grinding or lapping of duplicate curved surfaces, the preferred manner of carrying the invention into effect is to provide the crystal blank with one convex-spherical electrode surface and one substantially opticallyfiat electrode surface. A cross-sectional view through the middle of such an element is shown in Fig. 1. Here a designates the optically-fiat electrode surface and b the opposite convexspherical electrode surface. Optimum performance is achieved when the departure of the convex surface from optical flatness is substantially no more than two ten-thousandths of aninch (.0002) and substantially no less than one ten-thousandth of an inch (.0001). The same limits obtain where, as in Fig. '7, the element is of bio-convex spherical contour. Crystal elements whose curvature is greater or less than that above set forth will usually fail to exhibit the desired harmonicfrequency response, although in some cases the fundamental and a single, odd harmonic-frequency may be achieved.

In manufacturing the element of Fig. 1, the preferred practice is to start with a blank (of the later specified orientation) whose thickness is slightly greater than that required to achieve the desired fundamental thickness-mode fre quency. As in prior art thickness-mode crystals, the length and breadth dimensions are usually unimportant. One of the major surfaces of the blank is selected as a reference face, and this face (a, Figs. 1 and 2) is ground and lapped until it is substantially optically flat and the element is of the required thickness. By optically fiat, as this term is herein used, is meant approximately within two one-hundred-thousandths of an inch (.00002). This optically fiat surface (a) is then preferably given a finish to a degree which may be defined as one step removed from a polish.

The thickness of the blank at this stage in its manufacture will be understood to be uniform. The unfinished surface of the blank is then placed in contact with a flat surface having an abrasive film thereon. Force is applied, as with the fingers, adjacent its corner (as indicated by the arrows in Fig. 2), and the blank moved over the abrasive material with a circular or figure 8 movement. Since no force is applied to the center of the blank, and since the quartz is relatively flexible, the edges of the blank will be subjected to the greater abrasive action, whereby this surface (b, Fig. 1) will assume a convexspherical contour. Frequent tests at this stage should be made to determine the oscillating characteristics of the element. When the desired characteristics are achieved, the last treated surface (b) should preferably be given a finish similar to that given the other electrode surface a.

As previously mentioned, when a crystal blank cut in accordance with the Baldwin and Bokovoy disclosure is given a convex electrode surface and is otherwise finished in accordance with my above identified earlier invention, the finished element will exhibit a zero temperature coefficient of frequency when vibrated at its funda mental frequency, but will not exhibit this desired characteristic when vibrated at its third harmonic frequency.

My present invention is predicated upon my discovery that crystal blanks out within a certain very narrow range of angles of tilt, and given a convex electrode surface, will exhibit a zero temperature coefficient of frequency when vibrated at the third harmonic frequency. Peculiarly enough, this very narrow range of angles obtains substantially irrespective of the exact third harmonic frequency at which the desired low temperature-frequency characteristic obtains. Stated another way, the general rule for obtaining a low temperature-frequency characteristic in a crystal element (when said element is to be vibrated at its fundamental frequency), as stated by Baldwin and Bokovoy, is:

The higher the desired frequency the greater the angle of orientation required to achieve a desired temperature coefiicient.

But I have discovered that this rule does not obtain in a'crystal element when said element is vibrated at a multiple of the fundamental frequency dictated by its thickness dimension.

I am not now prepared to state definitely that there exists more than one very narrow range of angles within which a zero temperature-frequency characteristic may be obtained at ultra high (harmonic) frequencies. However, I have succeeded in achieving this desirable characteristic in numerous elements cut from crystal blanks whose electrode faces are normal to a 25 W angle and inclined from 34 23' to 35 30' away from parallelism with the Z (optic) axis, toward parallelism with a minor apex face of the mother crystal, and I reveal this information with the expectation that others skilled in the art may extend my discovery to crystals tilted about other axes in either the major or minor direction.

The convention which I shall employ in describing my invention is that established by Baldwin and Bokovoy in describing their socalled V-cut crystals. The use of this convention is recommended for the reason that it is applicable, without change, to quartz of both right hand and left hand crystalline structure. In accordance with this convention, the major and minor apex surfaces of the mother crystal are employed as reference planes and certain W and Y-l-G axes are employed as reference axes. It is, therefore, first necessary to identify these planes and axes.

Referring to Figs. 3 and 4 of the drawing and having in mind that all unbroken quartz crystals terminate in a hexagonal pyramid, it will be seen that certain terminal surfaces of the quartz extend to the apex of the pyramid. These surfaces are designated M and are the major apex surfaces. Those terminal surfaces which do not touch the apex are designated N and are the minor apex surfaces of the mother crystal. Occasionally, a mother crystal will be found in which more than three of the apex or cap faces extend to the top of the pyramid; other crystals may have their pyramid ends broken off. No confusion, however, need exist as to the virtual location of the major and minor apex faces of a broken or otherwise abnormal crystal providing that the side faces, m and 1L, or one of them, is intact, for it will be apparent from an inspection of Fig. 4 that those side edges of the mother crystal which approach each other in the direction of its ends terminate in a major apex face, while those which diverge in this direction terminate in a minor apex face. This is so in the case of both left hand and right hand quartz.

Fig. 3 is marked to show an X (electric) axis and an adjacent Y (mechanical or crystallographic) axis. The Z (optic) axis marked in Fig. 4 is perpendicular to the plane of projection in Fig. 3. The W axes lie between an X axis and an adjacent Y axis in the X--Y plane, i. e., in a plane normal to the Z axis. In Fig. 3 but one W axis is marked. It forms a W-angle of 25 with that X axis which is designated X-X and is the particular W-axis necessary to be identified in carrying my present invention into effect. That axis which is normal tothe above identified 25 W-axis is designated Y+9. It is about this Y+9 axis that the blanks employed in carrying my invention into effect, are tilted. The direction of tilt is toward a minor apex surface (in this case, surface N of the mother crystal; and the range of angles within which the blanks to be used in carrying my invention into effect are substantially from 34 23 to 35 30. Angles within this range are herein occasionally referred to as V-angles or V -angles.

Examples of 20 megacycle crystals Assuming now that an element capable of vibrating at 20 megacycles (20,000,000 cycles) per second and exhibiting a substantially exact zero temperature coefficient of frequency at that ultra high frequency, is required. Such an element requires a blank dimensioned to respond to a fundamental frequency of one-third of 20 megacycles or 6.66+megacycles. The angles required to be known are (a) W=25 and (b) V"=35. The W angle is by no means as critical as the V angle in that it may be varied as much as one-half of one degree in either direction without affecting the oscillating characteristics of the finished element.

With this information in mind, a section or slab Ill (Fig. 4) say one inch thick, is first sliced from the body of the mother crystal; the thickness dimension of this section is parallel to, and the length-breadth dimensions normal to the Z axis. A bar I2 is then cut from this section It! along the designated 25 W-axis, and then (referring to Fig. a blank I4 is cut whose top and bottom electrode faces have been rotated exactly 35 (the V angle) about the said Y+6 axis in a direction away from parallelism with the Z axis toward parallelism with the plane of that minor apex surface of the mother crystal which is designated N in Fig. 3. The thickness of this blank prior to being finished with a convex electrode surface in the manner previously described, should preferably be slightly greater than the thickness dictated by the constant (K), at this angle, for a 6.66+megacycle crystal. The thickness dimension as measured through the center of the finished element is determined by the following formula:

where t is the thickness dimension measured in mils of an inch,

K =67 (the thickness constant for a blank Whose W angle 25 and whose V angle 35) f=is the desired fundamental frequency, in

megacycles.

Therefore,

or, solved t=10.06 mils of an inch.

In tests with several 25 W angle, 35 V angle, crystals 10.+ mils of an inch thick, vibrated at the third harmonic frequency of 20 megacycles, I have found that they exhibit a substantially exactly zero temperature coefficient of frequency over a temperature range of 60 to +60 centigrade.

I have made other 20 megacycle crystal elements departing slightly from the substantially exact angles above mentioned with the following results:

(2) 25 12' W angle, 35 16' V angle. This crystal exhibited a temperature coefficient of frequency of 1.01 cycles per million, per degree centigrade, over a temperature range of -60 +60 C.

(3) 25 11' W angle, 34 23 V angle. This crystal exhibited a temperature coefficient of frequency of +2.33 cycles per million, per degree centigrade, over the same temperature range.

Examples of 7.5 megacycle crystals I have out other crystals finished with a convex electrode surface within the described narrow range of angles and of other thicknesses. Thus, I have cut crystal elements in agreement with the above formula to .a thickness of 26.8 mils of an inch. The fundamental frequency of such elements is 2.5 megacycles. Their exact orientation and operating characteristics when vibrated at 7.5 megacycles (the third harmonic) are given below:

(5) 25 12 W angle, 35 16 V angle. Temperature frequency characteristic (60 C.+60 C.) at 7.5 megacycles equal to 1.14 cycles per million, per degree centigrade.

(6) 25 W, 35 V angle. Temperature frequency characteristic at 7.5 megacycles was substantially exactly zero over a range of -60 C. to+60 C.

(7) 25 11' W angle, 34 23 V angle. Temperature frequency characteristic (60 CA -60 C.) at 7.5 magacycles equal to +3.05 cycles per million, per degree C.

Obviously (in view of the cited formula), my invention is not limited to plural frequency crystal elements of the exact dimensions and fre quencies herein given.

Although certain specific ways and means for accomplishing the objects of my invention have been set forth, it is to be understood that they have been given for the purpose of explaining the inventive concept and should not be construed as definite limits to the scope of the invention. Neither is it to be understood that any statements herein made in regard to dimensions, angles of orientation, etc, are other than close approximations. My invention, therefore, is not to be limited except insofar as is necessitated by the prior art and by the spirit of the appended claims.

What is claimed is:

1. A quartz piezo-electric element having a convex electrode surface and an orientation within substantially 1, for each angle, of a 25 W angle, 35 V angle crystal, said element b ing characterized by exhibiting a substantially zero temperature coefficient of frequency when vibrated at a frequency which is three times the fundamental frequency dictated by the maximum thickness of said element.

2. A quartz piezo-electric element adapted to vibrate at a fundamental frequency which is a function of its thickness dimension, said element having at least one convex electrode surface, the

departure of said surface from optical flatness being substantially no more than two ten-thousandths of an inch (.0002) whereby said element will respond to a second thickness-mode frequency which is harmonically related to said fundamental frequency, said element having an orientation substantially within 1 for each angle, of a 25 W angle, 35 V angle crystal, said element being characterized by exhibiting a substantially zero temperature coefficient of frequency when vibrated at its said second-mentioned thickness-mode frequency.

3. The invention as set forth in claim 2 wherein the V angle of said blank lies within the range of 34 23 to 35 30.

4. The invention as set forth in claim 2 Wherein the V angle of said blank is substantially exactly 35.

5. quartz crystal element having a convex electrode surface, a maximum thickness dimension of substantially ten mils of an inch (.01), and an orientation within substantially 1, for each angle, of a 25 W angle, 35 V angle crystal, said element being characterized by exhibiting a temperature coefficient of frequency of less than three cycles per million, per degree centigrade when vibrated at a frequency of substantially twenty million cycles (20 megacycles) per second.

6. The invention as set forth in claim 5 wherein the orientation of said element is substantially exactly 25 W angle, 35 V angle, whereby the temperature coefficient of frequency of said element at said frequency is substantially zero.

'7. A quartz crystal element having a convex electrode surface, a maximum thickness dimension of substantially twenty-six and eight-tenths mils of an inch (.0268") and an orientation within substantially 1", for each angle, of a 25 W angle, 35 V angle crystal, said element being characterized by exhibiting a temperature coefficient of frequency of no more than 3.5 cycles per million, per degree centigrade, when vibrated a frequency of substantially seven million five hundred thousand cycles (7.5 megacycles) per second.

8. The invention as set forth in claim '2 wherein the orientation of said element is substan ally exactly 25 W angle, 35 V angle, whereby the temperature coefficient of frequency of said element at said frequency is substantially zero.

HENRY W. N. HAVIK.

Images