Thin film resonator filter having at least one component...

Wave transmission lines and networks – Coupling networks – Electromechanical filter

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C333S189000, C310S312000

Reexamination Certificate

active

06377136

ABSTRACT:

FIELD OF THE INVENTION
The present invention relates to a thin film resonator, more particularly to a thin film resonator filter circuit.
DESCRIPTION OF THE RELATED ART
Thin film resonators (hereinafter “TFR”) are typically used in high-frequency environments ranging from several hundred megahertz (MHz) to several gigahertz (GHz).
FIG. 1
illustrates a side view of a conventional TFR component
100
. In
FIG. 1
, TFR component
100
includes a piezoelectric material
110
interposed between two conductive electrode layers
105
and
115
, with electrode layer
115
formed on a support structure
120
. The support structure
120
may be a membrane, or may be a plurality of alternating reflecting layers on a solid semiconductor substrate which may be made of silicon or quartz, for example. The piezoelectric material is preferably one selected from the group comprising at least ZnO, CdS and AlN. Electrode layers
105
and
115
are formed from a conductive material, preferably of Al, but may be formed from other conductors as well.
These TFR components are often used in filters, more particularly in TFR filter circuits applicable to a myriad of communication technologies. For example, TFR filter circuits may be employed in cellular, wireless and fiber-optic communications, as well as in computer or computer-related information-exchange or information-sharing systems.
The desire to render these increasingly complicated communication systems portable and even hand-held place significant demands on filtering technology, particularly in the context of increasingly crowded radio frequency resources. TFR filters must meet strict performance requirements which include: (a) being extremely robust, (b) being readily mass-produced and (c) being able to sharply increase performance to size ratio achievable in a frequency range extending into the gigahertz region. However, in addition to meeting these requirements, there is a need for low passband insertion loss simultaneously coupled with demand for a relatively large stopband attenuation. Moreover, some of the typical applications noted above for these TFR filters require passband widths up to 4% of the center frequency (for example, for a center frequency of 2 GHz, the bandwidth required would be about 80 MHz. This is not easily achieved using common piezoelectrics such as AlN, especially in combination with a plurality of reflecting layers on a solidly mounted substrate.
A conventional electrical circuit model for these resonators is illustrated in FIG.
2
A. The circuit model is a Butterworth-Van Dyke model (BVD), and is comprised of a series RLC line which represents the motional (acoustic) resonance of the TFR between an input terminal
10
and output terminal
20
. The series RLC line is in parallel with a capacitor C
O
representing the parallel plate capacitance (static capacitance) of the electrodes (for example, electrodes
105
and
115
in
FIG. 1
)
Impedance analysis of the BVD model illustrated in
FIG. 2A
yields a set of two resonant frequencies, a zero resonant frequency (“zero”) followed by a pole resonant frequency (“pole”). The separation of the zero from the pole is dependent on a piezoelectric acoustic coupling coefficient known as K
2
. This coefficient is a measure of how much of the acoustic energy is coupled into electrical, and varies with the piezoelectric material used in the TFR. As will be explained later, attempting to emulate standard LC bandpass filter design techniques is difficult because each TFR in the filter has the two resonances (pole and zero), whereas there is only one resonance in standard LC bandpass filter branches for common filter responses such as a Chebychev response, for example. As will be discussed later, this “extra” resonance (either the pole or zero resonance, depending on the design of the TFR) is disadvantageous in that it somewhat interferes with the passband as wider passbands are attempted. However, if materials with a larger K
2
are used, this extra resonance will be farther away from the passband, which ultimately yields a wider bandwidth for the passband filter (the size of K
2
is directly related to the amount of separation between pole and zero resonances).
A standard approach to designing filters out of resonators is to arrange them in a ladder configuration in a series-shunt relationship (i.e., a “shunt” resonator connected in shunt at a terminal of a “series” resonator). Each of the shunt and series resonators has a pole resonance and a zero resonance. To achieve a bandpass filter response, it is necessary to shift the pole frequency of the shunt resonator down in frequency to align somewhat with the zero frequency of the series resonator. This shifting of shunt resonator pole frequencies down in an attempt to match the series resonator pole frequency is typically accomplished by adding some material (such as a metal, metal oxide, etc.) to the top electrode of the shunt resonator.
Currently, the conventional way of designing TFR ladder filters is to design simple building blocks of TFR components which are then concatenated together (connected or linked up in a series or chain). In a simplified view, concatenation helps to achieve a larger stopband attenuation for the overall filter because each individual linked up section in the chain successively filters the signal more as it passes through the chain.
FIGS. 2B and 2C
illustrate the simple building blocks, which are commonly known as T-Cells and L-Cells. Referring specifically to
FIG. 2B
, a T-Cell building block
125
includes three TFR components
130
A,
130
B and
135
. TFR's
130
A and
130
B are “series arm” portions of the T-Cell block, being connected in series between an input port
132
and node
136
, and node
136
to an output port
134
of T-Cell building block
125
respectively. Resonator element
135
comprises the “shunt leg” portion of T-Cell building block
125
, being connected in shunt between terminal
136
and ground. Similarly in
FIG. 2C
, an L-section block
145
used in a conventional TFR filter circuit includes TFR
146
comprising the series arm portion, with a TFR
147
connected in shunt to TFR
146
at terminal
144
to ground.
FIG. 3
illustrates a conventional TFR filter circuit. The filter circuit of
FIG. 3
is created by concatenation of four T-Cells
151
-
154
. As discussed above, the chaining up of a plurality of T-Cells provides a filter with high stopband attenuation. Further, redundant series resonators may be combined in order to reduce the size of the filter, as illustrated in FIG.
5
. In
FIG. 5
, redundant components are shown combined (as compared to
FIG. 3
, series arm TFR components
160
and
165
are effectively “pulled” (i.e., combined) together to form one TFR series component
175
) in an effort to save space. Although the “inner” series branch electrode capacitances (see COS
2
of TFR components
175
,
180
and
185
) in
FIG. 5
are now different than the outer electrode capacitances (see COS of TFR components
190
and
195
) because of the post-design combining, the “root” design is based on all of the series branch electrode capacitances being equal, and based on all the shunt branch electrode capacitances being equal, as illustrated in FIG.
3
.
However, designing the filters illustrated in
FIGS. 3 and 5
by the conventional concatenating approach has certain disadvantages. Namely, the filter designed by the above approach suffers poor flexibility in widening the passband width, as well as poor flatness performance when deviating from the center frequency.
FIGS. 4A and 4B
illustrate these effects in terms of insertion loss and return loss, as dB (y-axis) versus unit frequency (0.02 GHz/division, x-axis).
FIG. 4A
illustrates some common flatness and asymmetry problems.
FIG. 4B
illustrates common non-equiripple return loss.
FIGS. 6A and 6B
illustrate the effects of using the conventional concatenation approach to widen bandwidth, depicting these effects in terms of insertion loss and return loss, as dB versus unit frequency. In order to a

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Thin film resonator filter having at least one component... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Thin film resonator filter having at least one component..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Thin film resonator filter having at least one component... will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-2895438

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.