US3573672A

US3573672A - Crystal filter

Info

Publication number: US3573672A
Application number: US771843A
Authority: US
Inventors: Irvin E Fair; Edwin C Thompson
Original assignee: Bell Telephone Laboratories Inc
Current assignee: AT&T Corp
Priority date: 1968-10-30
Filing date: 1968-10-30
Publication date: 1971-04-06
Anticipated expiration: 1988-04-06
Also published as: NL6916084A; CH501340A; NL155693B; FR2021887A1; DE1953826C2; DE1953826A1; BE740969A; GB1287002A; SE365083B

Abstract

The coupling between two loosely coupled electrode pairs mounted on the same crystal body to form a monolithic filter is vernier adjusted by plating an area in the interelectrode region.

Description

United States Patent Inventors Irvin E. Fair Treasure Island, Fla.; Edwin C. Thompson, I-Iokendauqua, Pa. Appl. No, 771,843 Filed Oct. 30, 1968 Patented Apr. 6, 1971 Assignee Bell Telephone Laboratories, Incorporated Murray Hill, NJ.

CRYSTAL FILTER 3 Claims, 20 Drawing Figs.

[1.8. CI 333/72, 3 10/ 8.3 Int. Cl l-l04r17/00, I-I0l r 7/00 Field of Search 333/70, 72; 310/83 Primary Examiner-Eli Lieberman Assistant Examiner- Tim Veazeau Attorneys-R. .I. Guenther and Edwin B. Cave ABSTRACT: The coupling between two loosely coupled electrode pairs mounted on the same crystal body to form a monolithic filter is vemier adjusted by plating an area in the interelectrode region. 7

Patented April 6, 1971 3,573,672

6 Sheets-Sheet 1 I F/G./

/N l/EN TORS 1'. E. FA/R EC. THOMPSON [an Law A T TORNEV Patented April 6, 1971 6 Sheets-Sheet 2 FIG. 6

Patented April 6, 1971 3,573,672

6 Sheets-Sheet 5 INSERTION LOSS lNdb Patented April 6, 1971 FREQUENCY SEPARATION -3 FREQUENCY SEPARATION f f I N KHZ AT l0 MHZ 6 Sheets-Sheet 4- F/G. l2

PLATER I0 [/24 /I4 FREQUENCY j GENERATOR I I2 I8 C16 METER N4.0" I ao- 2.0-

53 I ELECTRODES, ALIGNED ALONG z AXIS l I I I l l I I I I o I 234 567 89l0|ll2 i ELECTRODE SEPARATION t CRYSTAL BODY THICKNESS FIG. /4

l0- e.o '=2.a9

3.4 4.0- 3 I 3.0- =4.29 -%=5.24 015.29 |.o- ELECTRODES ALIGNED E 0.8- ALONG z'AxIs =Is.25 0 0T5 I.'o I15 210 2'5 3'.o

PERCENT PLATE- BACK loo (f-f Patented April 6, 1971 6 Sheets-Sheet 5 NHZ FREQUENCY OF f f I I j CRYSTAL PLATE& WAFER THICKNESS afi) I B A APPROXIMATES zAxls ELECTRODE ALIGNMENT 3?:2O'MAX 0 I I2 :0 I 0 7 s I; i ELECTRODE SEPARATION t CRYSTAL BODY THICKNESS IIIIIIIHW IIIIIIIIIIIllI/I I =I2 TYPICAL Fla. /5 HM MIDBAND FREQUENCY I400 I000 BANDWI DTH f f BANDWIDTH IN HZ PLATE-BACK OF DEPOSITION IN PERCENT X I0 CRYSTAL FILTER REFERENCE TO COPENDING RELATED APPLICATIONS This application relates to the copending applications, Ser. No. 541,549, filed Apr. 11, 1966 now abandon and Ser. No. 558,338, filed Jun; 17, I966, both ofW. D. Beaver and R. A. Sykes, assigned to the assignee of the present invention. The application also relates to the application of R. L. Reynolds and R. A. Sykes Ser. No. 723,676, filed Apr. 24, 1968, and also assigned to the same assignee as this application. The subject matter of these applications is herewith incorporated as part of this application as if recited herein.

BACKGROUND OF THE INVENTION This invention relates to energy transfer devices and particularly to crystal filters.

According to the beforementioned applications low-loss transmission of energy through an acoustically resonant crystal wafer vibrating in the thickness shear mode is selectively controlled by covering the opposite faces of the wafer with a number of spaced electrode pairs whose masses are sufficient to concentrate the thickness shear vibrations between the electrodes of each pair so that the pairs form separate resonators with the crystal, and by spacing the pairs far enough so that the coupling between any two adjacent resonators is less than a given amount.

According to an aspect of the beforementioned application, these capabilities may be exploited to form a filter that controls the passband between an electric source and a resistive load. This is accomplished by vapor depositing two or more pairs of electrodes on opposite faces of a piezoelectric crystal wafer. When one pair is connected to a source capable of exciting thickness shear mode vibration in the wafer, and when another pair is connected to a resistive load, the pairs form successive resonators with the wafer. The passband at the load can be adjustably controlled by varying the masses of the electrodes and the spacing between the respective resonators. Specifically, it requires making the electrodes sufficiently massive and spacing them far enough apart so that the coupling between adjacent resonators is at least small enough to be in what is called herein the controlled-coupling condition. Resonators in this condition have also been called definitively coupled.

The controlled-coupling condition becomes evident when the coupling between any adjacent two of the resonators is small enough so that the two short circuit resonant frequencies exhibited by these resonators are close enough in frequency to exclude the antiresonant frequencies caused by each resonator, from between them. The controlled-coupling condition is characterized by a characteristic impedance whose real portion varies with rising frequency from zero to a finite intermediate peak and returns to zero. For determining this condition the two resonators are decoupled from any other resonators.

In such filters the resonator couplings, whose value and number determine the resulting passbands, depend upon many factors, such as mass loading, electrode separation crystal wafer thickness, and electrode dimensions. Photoetching of plating masks and accurate crystal thickness permit reasonably good control over the couplings and resulting passbands. However, variations from the desired passbancb do exist. Overcoming such variations after the crystal structure is made involves adjustment of the masses of the electrodes in the separate resonators. However, this change in mass loading varies not only the couplings from resonator to resonator, but also the midband frequencies. As a result, while the overall passband and midband frequencies can be controlled within comparatively coarse limits, fine control is difficult.

THE INVENTION According to the invention, these difficulties are obviated by changing the crystal thickness in the region between the electrode pairs. Specifically, this is done by depositing additional material upon the wafer, or removing material from the wafer, between the pairs.

According to another feature of the invention the material is deposited or removed from one side of the wafer between the electrodes of each coupled pair of electrodes to be adjusted.

The invention is based in part on the recognition that chang ing the thickness of the region between pairs, just as changing the mass loading, affects both the bandwidth and the midband frequency. However, the rates of change are sufficiently different to permit fine adjustment by alternating the changes.

According to another feature of the invention the deposit or removal is in the shape of a spot, or preferably, a strip.

According to yet another feature of the invention the mass of the deposition or removal is regulated alternately with changes in the mass loading.

These and other features of the invention are pointed out in the claims. Other objects and advantages of the invention will become known from the following detailed description when read in light of the accompanying drawings.

FIG. I is a partly pictorial schematic diagram of a filter circuit, including a plan view of a filter, embodying features of the invention;

FIG. 2 is a diagram of the same circuit in FIG. I but showing a cross section of the filter in FIG. I;

FIG. 3 is a diagram of a filter circuit utilizing a monolithic crystal filter having two coupled resonators;

FIGS. 4 and 5 are, respectively, the lattice and ladder equivalent circuits for the filter circuit in FIG. 3;

FIG. 6 is a graph illustrating the variation in impedance and reactance with frequency of impedances in FIG. 4 when electrodes in the circuit of FIG. 3 are substantially massless and the coupling between electrode pairs tight;

FIG. 7 is a graph illustrating the image impedance, i.e. characteristic impedance, of the circuit of FIGS. 3, 4 and 5 for the conditions illustrated in FIG. 6;

FIG. 8 is a graph illustrating the transmission characteristics of the circuit in FIG. 3 when operated under the conditions of FIGS. 6 and 7 when terminated by a low resistance;

FIG. 9 is a graph illustrating the variations of impedance and reactance of elements in the circuit of FIG. 4 when the electrode pairs in the filter of FIG. 3 are coupled less than a predetermined amount;

FIG. 10 is a graph illustrating the image or characteristic impedance of the filter in FIG. 3 for variations in frequency under the conditions illustrated in FIG. 9;

FIG. I l is a graph illustrating the transmission characteristic of the filter of FIG. 3 operating under the conditions of FIGS. 9 and 10 when terminated by a low resistance;

FIG. 12 is a circuit diagram illustrating a test circuit for testing the coupling between resonators, that is electrode pairs, on a filter embodying features of the invention;

FIGS. 13, 14 and I5 are graphs illustrating the parameter relationships for filters such as those of FIGS. 1, 2, 3, 17, I8 and 19 without their interelectrode depositions;

FIG. 16 is a graph illustrating the effects of the deposited interelectrode area according to the invention upon the bandwidth and coupling of electrode pairs in FIGS. 1 and 2 and as measured in FIG. 12;

FIG. I7 is a diagram illustrating another filter embodying.

features of the invention;

FIG. 18 is a circuit diagram including another filter embodying features of the invention shown in plan view;

FIG. I) is a circuit diagram of FIG. 18 wherein the filter is shown as cross section 19-19 on FIG. I8; and

FIG. 20 is a partially pictorial circuit diagram illustrating another embodiment of the invention.

In FIGS. 1 and 2, two pairs of

electrodes

10, 12 and 14, 16 each have their constituent opposing electrodes vapordeposited or otherwise plated on opposite faces of a rectangular AT-cut quartz crystal wafer 18. The pairs, made up of rectangular electrodes whose thicknesses appear exaggerated for clarity, are aligned in this embodiment along the Z crystallographic axis of the crystal wafer. Through the leads 20 a high frequency source 22 applies a high frequency potential across the

electrodes

10 and 12 for piezoelectrically generating thickness shear mode vibrations in wafer 18. The portion of the vibratory energy in the wafer 18 between the

electrodes

14 and 16 establishes a varying electric field that the leads 20 apply across a terminating resistor R The two electrode pairs thus form two coupled resonators. A deposition 24 in the shape of a spot, vapor-deposited or otherwise plated between the electrodes and 14 on one side and the wafer 18 in the center of the interelectrode spacing, has smaller dimensions than the rectangular electrodes. Its mass is sufficient to widen the bandwidth of the filter to a desired value.

Disregarding the deposition 24, the masses of the

electrodes

10, 12, 14 and 16 are sufficiently great and the respective electrode pairs 10, 12 and l4, 16 are spaced from each other so that the resonators formed by the electrode pairs are in what is here termed the controlled-coupling condition. That is to say, the masses of the

electrodes

10, 12, 14 and 16 are sufficiently great so as to trap" or concentrate the energy of vibrations in the wafer 18 to the volume of the wafer between the electrodes of each pair and attenuate the energy exponentially with distance away from the pair. This limits the effect of the wafer boundaries upon vibrations within the wafer. At the same time, to achieve the controlled-coupling condition, the spacing between the electrode pairs combined with the degree of mass loading is such as to couple the pairs only enough so that resonant frequencies f and f exhibited by the coupled resonators are close enough so that neither antiresonant frequency f,,,, nor f exhibited by the respective resonators appears between them. More specifically, the coupled resonators are coupled to less than one-half of the maximum coupling in the controlled coupled condition. That is, the resonant frequencies are closer to each other than to the nearest antiresonant frequency.

The effects of having only two such electrode pairs can be appreciated by considering a filter such as shown in FIG. 3, its lattice equivalent circuit in FIG. 4 and its ladder equivalent circuit in FIG. 5. FIG. 3 corresponds to FIGS. 1 and 2 without deposition 24. In the ladder equivalent network the three capacitors C, represent the electrical equivalent of the acoustical coupling between the electrode regions of FIG. 3. According to Bartletts bisection theorem the two circuits are related to each other by the following equations:

C1 1B-1 g The values C and L are such that the shear mode fundamental frequency of the crystal wafer 18 is /zrnLQ. The value of L is a function of the crystal body thickness and the geometry of

electrodes

10, 12, and 14, 16. C is the static interelectrode capacitance of each pair of electrodes.

In FIG. 3 the signal transferred by the structure is greatest, and hence the insertion loss is least, when the characteristic impedance, i.e. the image impedance, Z,=R Thus maximum signal transfer and minimum insertion loss occur at those frequencies when Z, exhibits resistive values R i.e. when it is real and positive, so that Z,=R,=R Generally, the image impedance, i.e. the characteristic impedance Z,=1 Z Z where cuited, and Z is the input impedance when the load end is short circuited. Thus the characteristic impedance or image impedance Z, for the crystal structure of FIG. 3 and its equivalent circuit in FIG. 4 is equal IOVZAZB- Since the crystal wafer 18 has a large Q, the values Z and 2,, are almost exclusively comprised of their component reactances X A and X,,.

Thus the characteristic impedance Z, is substantially equal to vX X The values X A and X B can be plotted and the valuesof Z,- determined therefrom for various masses of

electrodes

10, 12 and 14, 16.

In the crystal structure of FIG. 3 when the wafer 18 is insignificantly mass loaded by the electrodes 10, l2, l4 and 16, vibratory energy generated between the

electrodes

10 and 12 decreases only gradually in other parts of wafer 18. Thus the wafer couples the electrode pairs tightly. The reactances X and X B of the impedances Z,, and 2,, then vary with frequency as shown in FIG. 6.

Since X A and X are imaginary numbers, VX X is real only if X A and X bear opposite signs. Thus, in the frequency regions in which X A and X appear on opposite sides of the abscissa of FIG. 6, the filter exhibits real positive characteristic impedances Z,==R,. 'As shown in the graph of real Z i.e. R,-, in FIG. 7 two real positive image impedances Z, exist for the type of coupling in FIG. 6. They extend, respectively, across the lower resonant-to-antiresonant range f,, to f,, and the upper resonant-to-antiresonant range f to fl of the individual impedances Z,, and Z The widths of these ranges are approximately equal and a function of the wafers piezoelectric coupling and the electrode areas. Since the insertion loss is minimum when the terminating impedance R of FIG. 7 matches the real characteristic impedance R,-, the insertion loss for such device is very high in the reactive impedance regionf to f It is low only at the two frequencies where R intersects R,-. Resistance R no matter what its value, intersects R,- of FIG. 7 in two places. Thus the curves of FIGS. 6 and 7 produce the insertion loss or transmission characteristic shown in FIG. 8. For any value of R this results in two minima separated by a wide band of loss.

According to the copending applications mentioned before, giving the electrodes sufficient mass concentrates the thickness shear mode vibration energy in the wafer 18 between the electrodes of the respective pairs so that the wafer 18 vibrates with greatly diminishing amplitude outside the volume between the electrodes. The coupling between the resonators decreases. Significant energy is not permitted to reach the boundaries of the wafer. Such mass loading of the electrodes produces two resonators when two pairs of electrodes are used. When these two resonators are placed in each others effective field, they operate similar to a tuned transformer.

Decreasing the distances between the electrode pairs and increasing the masses of the electrode pairs regulates the band spectrum through which the energy of the system of one pair passes through the system of the other pair. When this happens the resonant frequencies f,, and f approach each other. When the coupling is low enough so that f is less than f the individual reactance curves X,, and X appear as in FIG. 9. There, the individual resonant-to-antiresonant ranges of X A and X B overlap. Otherwise stated, f -f fi, j: There is no f or f,, between f and f,,. The resulting real portion of the image impedance Z, that is R,-, appears in the real plane of FIG. 10. As shown in FIG. 10 the impedance Z, possesses two positive real ranges. One range extends between the resonant frequencies j} and f,; and has an intermediate maximum R with zero extremes. A second range lies between f and f,,,,. There R, starts at infinity, drops and returns to infinity as the frequency rises.

One of the two frequency ranges of FIG. 9 can be rejected by terminating the

electrodes

14 and 16 within the resistance range of one resistance R,- but remote from the other. Since in FIG. 10 R closely matches all resistances less than Z the system passes the frequencies between f,, and f,, with little loss.

Z is the input impedance when the load end is open-cir- A curve showing the insertion loss for a filter exhibiting these f,,faA fl, fi,, is known as the beforementioned controlledcoupling condition. If f,,f exceeds or is equal to f,, ,-f,,, the condit ons of FIGS. 6, 7 and 8 exist. The coupling coefficient k between these pairs is equal to (f,,f,)/ fBf,. Approximately 15 For practical purposes, in order to make the maximum characteristic resistance value of FIG. 10 between f and f much smaller than the minimum characteristic impedance value between f g flu, the frequency difference f,,f, is generally below (f f,,,,)/2. Thus f, and f,; are closer to each other than to either f or f This assures adequate rejection of one band and passage of the other with suitable terminating values of R In FIGS. land 2, the electrodes l0, l2, l4 and 16, are in the controlled-coupling condition where f,f,, f,, ,f,,. That is, they follow the rule illustrated in FIGS. 9, 10 and 11. Nof or f exists between f,, and f More specifically, they are such that f,,f,, (f ,,f,, )/2. Thus, f,, and f,; are closer together than to either f,,,, or f,,,,. This is so both before and after the spot electrode 24 is added.

The bandwidth (fl -f of such a filter is a function of several parameters. The graphs of FIGS. 13, 14 and 15 illustrate empirical relationships between the parameters in one such filter. In these graphs the masses of the electrodes are represented not directly, but by how much the masses lower the frequency of each resonator. Such frequency lowering occurs even for a single pair of electrodes on a crystal wafer. The fractional drop (ff,.)/f in the resonant frequency f,, of an uncoupled resonator formed by a single pair of electrodes on a crystal wafer, from the fundamental thickness shear frequency f of the unelectroded crystal body, due to increasing masses of the electrodes is called plateback.

Adjusting the bandwidth (f,,f,,) of such filters has hitherto been accomplished by adding or removing mass from the respective electrodes. Adding mass tends to reduce coupling between resonators. Adding mass thus moves f and f close together, while removing mass separates them. However, adding the mass also lowers f and f Each absolute decrease in bandwidth is accompanied by a far larger absolute drop in prevailing midband frequency f or (f +f )/2. For example, as shown in FIG. 14, decreasing a 2 kHz. bandwidth at 10 MHZ. to 1.8 kHz. requires a change in plateback from 2.0 percent to 2.1 percent. This constitutes 0.1 percent change in f,, at 10 MHz. The 200 Hz bandwidth drop is accompanied by a 10 kHz. drop in the frequency of f,,, f,; and f As a result, the dimensions of the dual-mode resonator in FIG. 3 have to be accurately determined beforehand to achieve an accurate passband at the desired midband frequency. For example, the thickness of the wafer determines the fundamental thickness shear frequency f of the wafer for any particular axial alignment of electrodes.

However, according to FIGS. 1 and 2 the deposition 24 is used to control the band. As the mass of the deposition 24 increases, the absolute changes in bandwidth and center frequency f in Hz are comparable. However the relative frequency change is small for a large bandwidth change. If the small relative frequency change is too large for the filter tolerances, the midband frequencyf may be adjusted back by mass loading the electrodes. When the electrodes are mass loaded, in contrast to adding to the deposition 24, the absolute change in midband frequency f is accompanied by almost insignificant changes in bandwidth.

FIG. 16 illustrates the increase in bandwidth f,,f as a result of depositing and increasing the mass of the deposition 24 as measured in the circuit shown in FIG. 12. To obtain the graph, two gold electrodes 0.200 inches by 0. l 20 inches with a 0.l00 inch separation between the longer edges were oriented along the Z axis of a l5-millimeter square quartz I crystal wafer having a fundamental frequency of 8 MHz. Sufficient mass was given the electrodes to produce a two percent plateback, that is, (f-fr)/f=0.02. Before adding the spot shaped deposition 24. the frequency generator 30 applied energy through the measuring resistor 34 to the

electrodes

10 and 12. The voltmeter 34 measured the phase angle across the resistor 32. The

electrodes

14 and 16 were short-circuited. The plater 36 then applied gold into the interelectrode space to form the deposition 24.

Since no connections were made to the deposition 24, the mass loading of the spot-shaped deposition 24 was determined on the basis of plating time. This determination was accomplished by calibrating the amount of mass loading obtained on a similar resonator with the same electrode geometry. Gold was deposited on the electrode at various time intervals and the corresponding bandwidth as well as midband frequency change noted.

The percent increase in bandwidth was measured at four intervals, at approximate mass loadings of 0.024, 0.035, 0.064 and 0.077 percent. The percentage increase in bandwidth as shown in FIG. 16 was approximately linear. At the same time the short circuit frequencies of resonators f, and f were measured to establish (fl f and (f fO/Z. At the start of measurements the bandwidth of the filter as determined by the frequency difference between the frequencies f, and f was 986 Hz. As mass was added to the deposition 24, the bandwidth increased in steps until a bandwidth of H34 Hz was achieved for an equivalent mass loading of the plated area of 0.077 percent. This is an increase over the initial bandwidth of 148 cycles, or more than 15 percent. The value (f,,+f.)/2, which is the midband frequency f exhibited by the filter, decreased 193 cycles, or on the order of 0.0022 percent. The absolute increase in bandwidth approximated the absolute decrease in midband frequency f The midband frequency could be returned to its original value by removal of a small amount of plating from the electrodes. This would result in a bandwidth change of only two or three Hz, or 0.4 percent.

Manufacture of a filter such as shown in FIGS. 1 and 2 is accomplished by etching a crystal wafer and cutting masks to dimensions determined by the graphs of FIGS. l3, l4, and 15. Gold is then vapor deposited onto the crystal wafer so that the exhibited midband frequency f =(f +f )/2 is slightly higher than the desired midband frequency f,,, and the bandwidth (f,,f,,) equal to or less than the desired bandwidth.

The coupling, prevailing midband frequency f and band width are measured as shown in FIG. 12 by applying signals from a frequency generator 30 across the

electrodes

10 and 12 and short circuiting the

electrodes

14 and 16. The voltmeter 32 measures the phase angle across a measuring resistor 34. Maximum voltages measured by the meter 32 indicate the frequencies f,, and f The difference should be less than f,,,,-f so that the resonators are in controlled-coupling condition.

If the bandwidth f f is less than desired, the plater 36 applies gold to form the deposition 24 as shown in FIG. 12. As more gold is applied to the deposition 24, the bandwidth increases. For example, it may increase a kHz. bandwidth 150 Hz or I5 percent. At the same time the midband or center frequency f or (f -l-f )/2 decreases a comparable number of Hz. This is a small proportion of the center frequency and may be insignificant. When the desired bandwidth is reached, the filter is completed if the accompanying midband frequency drop is within desired tolerances. If not, additional mass is added to the

electrodes

10, 12 and 14, 16 to shift the existing midband frequency (f +fA)/2 to the desired midband frequenc f,,,.

This added plateback changes the bandwidth only an insignificant amount because, although the absolute lowering in the number of cycles may be large, it represents but a slight proportionate change in the midband frequency. Such slight changes produce only slight changes in the proportion of the bandwidth and result in even slighter absolute bandwidth changes. As can be seen from FIG. 14 lowering the midband frequency f,, 150 Hz in l MHz, or changing the plateback l00 l 50/1 0=0.0015 percent. may change the bandwidth approximately 4 Hz. The invention thus takes advantage of the differences in the rates of change exhibited by the two methods of changing frequency and bandwidth.

According to another embodiment of the invention, as shown in FIG. 17, the deposition 24 is in the form of a rectangular strip. This configuration has the same effect as the spot in FIG. I. This configuration has the advantage of permitting simplified calculations of its effect.

FIGS. 18 and 19 illustrate another filter embodying features of the invention. Here, six pairs of

electrodes

40, 42; 44, 46; 48, 50; 52, 53; 56, 58; 60, 62 are deposited on an AT-cut quartz crystal wafer 64 along the Z axis. A source 66 supplies energy to the

electrode pair

40, 42 and a load 68 receives energy from the

electrodes

60, 62. The intermediate pairs of electrodes are short-circuited to each other and grounded. Each adjacent pair of electrodes is in controlled-coupling condition. That is, when other resonators are detuned, any two adjacent resonators exhibit the conditions of FIGS. 9, and 11. More specifically, with other resonators detuned, any two adjacent resonators exhibit the condition f,,j" (f,, -f 4 )/2.

According to the present invention, the strip-shaped deposition 72 in the interelectrode spacing between the

electrodes

40, 44, 48, 52, 56, and 60 compensates for departures from the required electrode dimensions, wafer thickness, and mass loading. Thus, where such departures may otherwise have resulted in unreliable bandwidths, couplings or midband frequencies, the existence of the depositions permits fine adjustment of the bandwidths and subsequent fine adjustment of the midband frequencies with negligible disturbance of the bandwidths. It requires only that the electrodes 40 through 62 be plated back initially less than required. The plates thus introduce vemier adjustments which can be added as necessary between any electrode pairs.

The term thickness shear mode is used as defined in Mc- Graw Hill Encyclopedia of Science and Technology, 1966, Vol. 10, pages 221 et seq. It includes both parallel face motion and circular face motion about a common axis. The latter is sometimes called the thickness twist mode.

The crystal structure of FIGS. 18 and 19 is manufactured by first selecting the coupling approximately to (f f,,)/f,, or (f -fn/f between each adjacent electrode pair about a desired midband frequency f,, on the basis of the bandwidths calculated for successive coupled resonators. The couplings are adjusted so that any error in bandwidth appears on the low side. An electrode size and suitable plateback (from 0.3 to 3 percent) are chosen from curves such as those in FIGS. l3, l4 and 15, which have been developed empirically. There 1 is the wafer thickness and r the width of the electrodes r/! is generally made equal to 12 although in practice any value between 6 and 20 is usable. A value of is often chosen as the length of the electrodes normal to the coupling axis for good suppression of other modes. The fundamental thickness shear mode frequency f is determined to correspond to the chosen plateback P,, from the formulas I PB=' I2' 1- PB (4:)

Here f,,, is slightly higher than the desired midband frequency f,, in order to make the midband frequency error appear on the high side.

The manufacture starts by first cutting a wafer 16 from a quartz crystal having the desired crystallographic orientation such as an AT-cut. The wafer is then lapped and etched to a thickness 1 corresponding to the desired fundamental shear mode index frequency f. either for parallel or twist motion. Generally, the thickness is inversely proportional to the desired frequency. Masks are placed on each face of the crystal wafer with cutouts for depositing the six electrodes. The geometry of the electrodes is determined by considering the desired bandwidths and the convenient plateback.

The proper separation d between the electrodes may be determined from the graphs such as those of FIGS. 13, 14 or 15 which show variations in percent bandwidth for various ratios of electrode separation to plate thickness and for various platebacks. as well as various values of r/t.

To obtain the chosen platebacks, gold or silver is deposited such as by vacuum vapor plating through the masks so as to make connections possible and achieve about nearly all of the total desired plateback. For this, energy is applied successively to each pair of electrodes while mass is added to the electrodes until the frequency shifts nearly to the desired frequency f,,,. The procedure is repeated for all the electrode pairs. During this procedure for each pair, the others remain opencircuited. However, it may be necessary to obviate the effect of the other pairs by terminating them inductively. The depositions 24 are added to achieve the desired individual couplings as necessary to obtain the desired bandwidth Bw.

Mass is then added to the plates 40 to 62 to plate them back further until each pair, when considered alone, resonates at f,,, and the midband frequency of the system is f,,.

According to still another embodiment of the invention the crystal material between electrode pairs is removed, by etching, for accomplishing the vemier adjustment. This is shown by the structure of FIG. 20. Here electrodes 80, as electrode pairs on a crystal wafer 82, form resonators in controlledcoupling condition. Recesses 84 between the electrode pairs serve to decrease the bandwidth while increasing the midband frequency f,, exhibited between any two adjacent pairs when they are considered alone. A deposition 86 corresponding to the deposition 24 serves the same adjusting function as the deposition 24. A source S which energizes the first electrode pair, thereby excites the crystal wafer 82. The last electrode pair energizes a load resistor R While embodiments of the invention have been described in detail, it will be obvious to those skilled in the art that the invention may be embodied otherwise without departing from its spirit and scope.

I claim:

1. An energy translating device comprising a crystal body, first plate means on said body for interacting with acoustical energy in said body, second plate means on said body and spaced from said first plate means for interacting with acoustical energy in said body, said plate means when said body is excited forming respective mutually coupled resonators together exhibiting interdependent resonant frequencies whereby an energy band may be translated between said resonators, said body having a continuous mass, and material means deposited between said plate means thereby altering the mass of said body.

2. An energy translating device comprising a crystal body, first plate means on said body for interacting with acoustical energy in said body, second plate means on said body and spaced from said first plate means for interacting with acoustical energy in said body, said plate means when said body is excited forming respective mutually coupled resonators together exhibiting interdependent resonant frequencies whereby an energy band may be translated between said resonators, said body having a continuous mass, said body including a wafer, said plate means each including a pair of electrodes on opposite faces of said wafer, said wafer comprising an AT-cut for vibrations in a thickness shear mode, and material means in the shape of a strip deposited on said wafer.

3. An energy translating device comprising a crystal body, first plate means on said body for interacting with acoustical resonators, said body having a continuous mass, and material means in the shape ofa strip deposited on said body thereby to alter the mass of said body whereby translated energy is conformed to a given energy band.

Claims

3. An energy translating device comprising a crystal body, first plate means on said body for interacting with acoustical energy in said body, second plate means on said body and spaced from said first plate means for interacting with acoustical energy in said body, said plate means when said body is excited forming respective mutually coupled resonators together exhibiting interdependent resonant frequencies whereby an energy band may be translated between said resonators, said body having a continuous mass, and material means in the shape of a strip deposited on said body thereby to alter the mass of said body whereby translated energy is conformed to a given energy band.