Application of Robert H. Cother, 437 F.2d 1399 (C.C.P.A. 1971)

This is an appeal from the decision of the Patent Office Board of Appeals affirming the rejection of claims 1-8 of appellant's application for a reissue patent.¹  No claims have been allowed.

All the claims were rejected on two grounds: (1) that new matter had been introduced into the reissue application, stated to be based on 35 U.S.C. § 251, and (2) that they were obvious in view of the prior art, under 35 U.S.C. § 103.

The invention relates to systems for amplifying the output of piezoelectric elements which generate electric charges. The piezoelectric elements may serve as the transducer in an accelerometer for study of the motion of vibrating objects. In such an accelerometer, a piezoelectric element having two opposite parallel faces may be mounted with one face secured to a housing that is placed on an object under investigation with the other face in contact with a mass member. When the object vibrates, the mass member tends to remain stationary so that the piezoelectric element is alternately compressed and expanded between the mass and the housing. Due to the element's piezoelectric characteristics, electric charges are developed on its opposite faces to provide electric voltages in accordance with the acceleration. The piezoelectric element is frequently connected across the input of an amplifier to provide amplified voltages for measuring the acceleration. Since the vibrations studied often involve components having frequencies that extend over a range from a few cycles per second to many thousand cycles per second, the amplifier should provide uniform amplification over such a range.

The application states that a piezoelectric accelerometer of the type which may be used to measure accelerations accurately over the aforementioned range "is shown, for example, in Patent No. 2,714,672" and that " [o]ther types of piezo-electric accelerometers that are suitable for such use are well known."

The application further states that a piezoelectric accelerometer inherently possesses a low capacitance such as 500 μμ.²  It is also stated that it is often desirable to locate piezoelectric transducers at points remote from the amplifiers and that the length of the connecting cable may vary by several hundred feet from one installation to another. As a result, the shunt capacitance of the cable (a typical value of which is stated to be 30 μμ. per foot) may also vary by a great amount from one installation to another, "thus affecting a great change in the cut-off frequency and a great variation of signal strength at all frequencies." The application then states:

According to this invention, the foregoing difficulties are overcome by employing an amplifier which utilizes a capacitive negative feedback circuit to render the input impedance of the amplifier capacitive over the range of frequencies of the signal components that are of interest.

In the best mode of practicing this invention now known, the input capacitance of the amplifier is made large compared with any changes that are likely to be encountered in the total effective source capacitance due to the use of cables of different lengths, and the amplifier input capacitance is very large compared with the capacitance of the piezo-electric transducer itself and any cable that is expected to be used.

An illustrative embodiment of the invention discloses the piezoelectric accelerometer connected to a three-stage charge amplifier which is provided with a negative feed-back circuit "employing a capacitor Cô for rendering the impedance looking into the amplifier * * capacitive over the range of frequencies of interest above the low frequency cutoff."

The appealed claims are directed to the combination of a signal source comprising a transducer, including a "piezoelectric ceramic element," with an amplifier in accordance with appellant's disclosure. Appealed claim 1, with additions and deletions with respect to original claim 1 in italics and brackets, respectively, is illustrative:

1. In combination: a signal source comprising a piezoelectric transducer that includes a piezoelectric ceramic element having a pair of relatively spaced apart relatively movable surfaces thereon, and having a pair of mutually insulated electrodes on said surfaces, said element and said electrodes forming a source capacitor having a capacitance C_a, [and] said source further comprising [a displaceable element] means for applying a force to said ceramic element to displace said surfaces relative to each other for developing a charge on said capacitor that varies in proportion to the relative displacement of said [element] surfaces, and an amplifier having an input connected across said electrodes to detect the charge thereon, said amplifier having an effective input capacitance C_e looking into said input, which capacitance C_e is large compared with the capacitance C_a of said source capacitor and the capacitance across the connections between said source and said amplifier, said amplifier also having a resistance looking into said amplifier input, said input capacitance C_e and said input resistance establishing for said amplifier a low cut-off frequency that is in the sub-audio range, the resistance in the connections between said source and said amplifier being low compared with the capacitive impedance of said connections and said source at frequencies in a predetermined range above said cut-off frequency, said amplifier including a negative feedback circuit connected between the output of said amplifier and the input thereof, said feedback circuit comprising feedback capacitor means having a capacitance C_o, the impedance of said feedback circuit having a capacitive component that is greater than the resistive component apparently in series therewith as measured across the ends of said feedback circuit at frequencies in said predetermined frequency range above said low cut-off frequency, the voltage gain A of said amplifier in the absence of feedback and the feedback ratio B established by said feedback circuit and said capacitance C_o of said feedback capacitor being so proportioned that the apparent capacitance C_e = (AB+1) C_o looking into the input of said amplifier from said source is large compared with the capacitance of said source so that the voltage output of said amplifier at any frequency within said predetermined range of frequencies is substantially proportional to the magnitude of the relative displacement of said electrodes [element] irrespective of the capacitance C_a of said source.

Claim 4, the only other independent claim on appeal, differs from claim 1 in additionally referring to the cable connecting the electrodes of the piezoelectric element of the transducer to the amplifier and to the capacitance of that cable.

Like claim 1, all the appealed claims are amended with respect to the patent claims to limit the signal source to one including a "ceramic" piezoelectric element. The disclosure in the patent does not include the word "ceramic" but states with respect to the illustrative embodiment:

The piezo-electric element * * * may be of any kind generally employed in accelerometers, such as barium titanate (BaTiO₄) elements or Rochelle or quartz crystals.

In the present application, the opening statement in the patent specification stating that the invention relates to sysems "for amplifying the output of charge-generating sources such as piezo-electric transducers" has been modified by adding the term "of the ceramic type." Also, the sentence quoted immediately above was changed to read:

The ceramic piezo-electric element * * * may be of any kind generally employed in accelerometers, such as barium titanate (BaTiO₄) elements.

Other similar changes were also made.

In the reissue oath, appellant asserts that "the patent is partly inoperative due to the fact that the claims recite more than what the applicant had a right to claim." He states also that the patent should have been directed to the use of ceramic piezoelectric elements and that the reference therein to "Rochelle or quartz crystals" is "inaccurate, and should have been deleted from the specification." He further states that "errors and deficiencies appearing in applicant's specification and claims" became apparent to him when Swiss patent No. 267,431 to Kistler³  came to his attention as the result of correspondence with prospective licensees under the patent.

The Swiss patent, which was not cited in the prosecution of appellant's patent, discloses an amplifier structure for use in measuring movements of machine parts by displacement of plates of a charged capacitor and variable forces by means of piezoelectric crystals. The patent emphasizes, as appellant also recognized, that the measured values in these cases are "represented by charges, not currents." It points out that amplification of voltages produced by such charges involves difficulties, including a tendency for the charge to be measured to leak off the grids of the amplifier tubes. It solves the difficulties in amplifying such charges by employing an amplifier circuit using an inverse or negative feedback capacitor between the output and input circuit, which capacitor is stated to have a "capacity [capacitance] * * great in comparison with the amplifier input capacity." The patent states that by use of this construction "the measuring range can be extended and a strict linearity as well as constant amplification can be achieved." The amplifier is disclosed as having either a single stage or multiple stages.

The rejection on new matter is grounded on the view of the examiner and the board that the change in the present specification to specify transducers of ceramic barium titanate and exclude "Rochelle or quartz crystals" and the change in the claims to specify "ceramic" piezoelectric elements are not supported by the patent disclosure. Appellant traverses that rejection with evidence which he contends shows that a person of ordinary skill in the art would have understood the patent disclosure to support the changes he embodied in the present application. While the "new matter" rejection is of interest as part of the background of the appeal, we find it unnecessary to determine its merits because the view we take of the rejection on the prior art makes it moot.

The prior art rejection is for obviousness under 35 U.S.C. § 103 in view of the combination of the Shoor⁴  patent and the Swiss patent described above. It was the view of the examiner and the board that it would have been obvious to substitute an amplifier constructed in accordance with the teachings of the Swiss patent in place of an amplifier of undisclosed construction which Shoor shows connected through a coaxial cable to an accelerometer including a barium titanate piezoelectric element.⁵ 

In traversing the rejection, appellant relies essentially on the limitation in the claims that the source or piezoelectric element is ceramic. That is to be expected, since it is primarily⁶  in that limitation that the appealed claims differ from the patent claims, which appellant asserted in his oath recite more than he had a right to claim.⁷  The argument is that ceramic piezoelectric transducers have a higher capacitance than quartz transducers and, " [a]ccordingly, the problems involved in amplifying the output of the ceramic transducer cannot be solved by merely following prior art suggestions for amplifying the output of quartz transducers."

Appellant urges, in support of his argument, that the Swiss patent "describes a quartz transducer connected to the input of a charge amplifier" (emphasis supplied), implying that the reference disclosure is limited to use of its amplifier with quartz crystals. That assessment is incorrect since the Swiss patent does not specify any limitation on the composition of the "piezo-electric" crystals which it discloses as producing charges that are to be amplified. In fact, the Swiss patent, as does appellant's patent, suggests broad use of its amplifier, including use to amplify charges from sources other than piezoelectric elements.

Appellant further attempts to lessen the significance of the Swiss patent by referring to an engineering report on an actual amplifier purportedly sold by the Swiss patentee in 1956 or 1957.⁸  It appears from a table of "Gain vs. Frequency Response" in the report that the device was subjected to input signals at a number of discrete frequencies from 0.5 Hz to 200, 50 and 20 kHz from sources of undisclosed structure having capacitances of 200 pf, 1000 pf, and 10,000 pf, respectively. The results might be described as revealing that the actual amplifier tested showed little variation in gain with frequency over the larger part of the ranges tested for source capacitances of 100 pf and 1000 pf, but considerably more variation where the source capacitance was 10,000 pf.⁹ 

This test report does not demonstrate that the Swiss patent would not have taught a person of ordinary skill in the art how to construct a charge amplifier which provided satisfactory amplification of charges produced by an accelerometer employing a piezoelectric element of ceramic barium titanate. At best, the actual amplifier was no more than an embodiment of the Swiss patent amplifier designed for use under some specific conditions undescribed by the report but obviously including treatment of the output of a charge source having a total capacitance for which the amplification characteristics were deemed satisfactory. No reason is revealed for doubting that a person skilled in the art applying the invention taught by the Swiss patent to different conditions, such as use with a source having a higher capacitance, would be able to obtain the "extended" measuring range, the "strict linearity" and the "constant amplification" which the patent states that its amplifier provides.

Accepting appellant's assertion that ceramic barium titanate piezoelectric elements have a higher capacitance than other piezoelectric elements such as quartz, we are still not satisfied that the invention claimed here solves any problem that a person skilled in the art could not solve from the teachings of the Swiss patent.¹⁰  Rather, we agree with the conclusions expressed by the board, with particular respect to independent claims 1 and 4, as follows:

We agree with the Examiner that it would be obvious to substitute the amplifier of the Swiss Patent (which is stated to be for use with a piezo-electric crystal) for the amplifier 22 of Shoor. Appellant's original disclosure refers to barium titanate, Rochelle crystals and quartz crystals as equivalents in connection with his amplifier. Although the Swiss Patent makes no mention of designing for a specific low frequency response, we take judicial notice of the fact that the equations for calculating the frequency response of an amplifier were well known long prior to appellant's original filing date. An example of this appears on pages 259-265 in the Fourth edition of "Electronic and Radio Engineering" by Terman published in 1955 by McGraw-Hill. The Swiss Patent clearly discloses the use of capacitive feed back to increase the effective input capacitance of the amplifier. It would be obvious to one of ordinary skill in the art to design the amplifier, in accordance with the impedance of the source, to obtain any desired low frequency characteristic.

With respect to claim 4, we further think it would be obvious that the cable which connects the piezoelectric element to the amplifier in Shoor would introduce additional capacitance into the total source capacitance that would require consideration in the design of the amplifier where the cable was of substantial length.

Claim 2 merely adds to claim 1 the recitation that the amplifier of the combination includes a plurality of stages which are direct current coupled. Since, as the board noted and appellant does not deny, the Swiss patent discloses such feature, the rejection of claim 2 is also sustained.

Claims 3 and 5 add to claims 1 and 4, respectively, recitations that the effective capacitance looking into the input of the amplifier exceeds 1.4 μf. In sustaining the rejection of these claims, the board stated:

According to appellant's own disclosure this is merely the calculated value to provide response [of the amplifier] down to 2 c. p. s. [cycles per second] and is merely a matter of design related to a specific impedance (not specified in either of the claims).

We find no error in the rejection of claims 3 and 5. As appellant concedes in his brief, the desirability of providing uniform response down to sub-audio frequencies as low as 1 to 5 cycles per second was "well known" prior to his invention. We think that a person skilled in the art would have found it obvious to design the amplifier of the Swiss patent for response at a frequency as low as 2 cycles per second for a particular source, including one having a high total source capacitance, and find no reason in the record for doubting his ability to do so.

Claims 6 and 8 are somewhat more specific concerning the capacitance of the connecting cable. The board considered a condition where the capacitance of the cable is greater than that of the transducer to be the well known and undesirable result of using a long cable and "certainly not unobvious." It further took the position that a person of ordinary skill in the art would use as long a cable as necessary. We agree with the board and are satisfied that a person skilled in the art would be aware that cable capacitance would be a factor which depends upon cable length and would require consideration as part of the total source capacitance where a long length was used. It thus would be obvious to such a person to consider the capacitance of the cable where it turned out to be greater than that of the transducer (claim 6) or where the length of the cable turned out to be greater than two hundred feet (claim 8).

Claim 7 adds to claim 4 a recitation that the amplifier includes a plurality of stages and that the input stage comprises a transistor. The Swiss patent discloses that its charge amplifier may employ multiple stages and the board was in agreement with the examiner that it would have been prima facie obvious to use transistors in that amplifier. Since appellant does not appear to challenge that conclusion before us, and in the absence of evidence of unexpected results from the substitution of transistors or other evidence of unobviousness, the rejection of claim 7 is sustained.

The decision is affirmed.

Affirmed.

BALDWIN, Judge (concurring).

I believe that the modifications appearing in this application over appellant's original disclosure and relied on to support the claims on appeal constitute new matter under 35 U.S.C. § 251 and I would affirm for that reason alone.