Surface acoustic wave filters

Surface Acoustic Wave filters , shortly SAW filter or SAW filter (of English surface acoustic wave called) are bandpass filters for electrical signals below 3 GHz with a low bandwidth of a few megahertz. The filters based on surface acoustic waves are mechanical filters .

Since their center frequencies can be significantly higher than the conventional quartz and ceramic oscillators, they are used as frequency-determining components in many receivers and transmitters, which are in the range beyond a few hundred megahertz (e.g. in the 433 MHz ISM band ) up to working several gigahertz, e.g. B. in radio data transmission, such as WLAN , or mobile communications .

AOW filters are also used as intermediate frequency filters in television receivers and cell phones . This technology is also widespread in radio remote controls for car keys.

AOW filter from a VHF / UHF tuner. Original size: 12 mm × 2 mm

Working principle

The above AOW filter under higher magnification

SAW filters are based on the interference of signals with different transit times , realized with the piezo effect . Two transducers ( interdigital transducers , or IDTs for short ), each consisting of a pair of electrodes that interlock in the shape of a comb (also known as fingers ), are applied to a piezoelectric single crystal . Due to the piezo effect and the electric field created on the fingers , the underlying crystal near the surface is distorted. The resulting deflections of the crystal lattice can overlap with a suitable excitation frequency due to constructive interference to form surface waves. Twice the distance between the individual fingers corresponds to the wavelength (lambda) of the surface acoustic wave with the desired transmission frequency (also center frequency) on the crystal. Together with the crystal-dependent velocity of the surface wave, the transmission frequency can also be determined. ${\ displaystyle \ lambda}$ ${\ displaystyle f_ {0}}$ ${\ displaystyle v_ {0}}$ ${\ displaystyle f_ {0} = {\ frac {v_ {0}} {\ lambda}}}$

The surface wave generated is converted back into an electrical signal by the second converter.

The structure forms an FIR filter . Therefore, through the dimensions of the crystal and the structure of the fingers (distance, shape, length, ...) the frequency response of the filter can be adjusted almost as desired.

When developing an AOW filter, it must also be ensured that the transmitter and receiver electrodes are neither inductively nor capacitively coupled. The temperature sensitivity is also disadvantageous.

The so-called " dart cells " and " Hunsinger cells " are among the most important types of finger structure .

Schematic representation

Structure of an AOW filter; the acoustic wave travels from left to right; The shape and scaling of the metallic fingers determine the transfer behavior

A reflector is then preferable if the filter has a low insertion loss (engl. Insertion loss ) is supposed to have. The reflector and its position must be adjusted to the center frequency so that the reflected and emitted waves do not cancel each other out at the center frequency, but add up. This reduces the losses when converting electrical signals into acoustic signals, as it were through "recycling".

The insertion loss of an AOW filter depends on the bandwidth. AOW filters with a very small bandwidth achieve an insertion loss of approx. 0.5–3 dB ; broader bands, however, only approx. 5–9 dB. This is because the bandwidth is inversely proportional to the number of fingers: large bandwidth → few fingers and vice versa. The fewer fingers are available, the less electrical signal power can be coupled in or out.

realization

SAW filters are built up photolithographically on the surface of piezoelectric single crystals. The crystal material used is, for. B. lithium niobate (LiNbO ₃ ) or quartz (SiO ₂ ). The shape and shape of the fingers consisting of a metal layer can be calculated or simulated with powerful computer programs in order to realize a certain transfer function.

Important properties of an AOW filter are insertion loss and quality factor . Further effects are, for example, the undesired generation of harmonics, also known as harmonics , which is dependent on the manufacturing process. They arise, for example, from large signal amplitudes, mechanical inaccuracies (adhesion, metal edges, contacting) and non-linear material properties.

Depending on the type of filter, harmonics that are electrically excited from the outside may or may not be desired. Multiband filters, for example, produce by multiple keying of the signal on the harmonics of several maxima, which transmit the same signal energy as the fundamental wave at . Such behavior is undesirable in the case of frequency filters, e.g. B. from mobile telephones , where only in a narrow sub-range of a frequency band ( rented by the network operator ) may be sent and received. ${\ displaystyle f_ {0}}$ ${\ displaystyle f_ {0}}$

Advantages and disadvantages

SAW filters are excellent bandpass filters , as they have a very strong suppression above and below the respective cut-off frequencies. In recent years they have undergone rapid development, with the result that almost any bandwidth , center frequency and attenuation can be produced.

High signal powers (> 10 W / + 40 dBm ) cannot be transmitted via an AOW filter. The conventional filters, which are discretely made up of capacitances and inductances, line circuits or cavity resonators, must be used here.

In the pass band, ripples in the transfer function are characteristic, which are higher the steeper the flanks of the pass band are. The causes are due to the principle and come from the superposition of the different frequency waves in the filter. They can be reduced by more and more sophisticated design, but in addition to transmission and blocking attenuation, they are an essential characteristic for characterization and are therefore specified in the manufacturer's specifications.

SAW filters show non-linear effects, especially with large signal amplitudes, which lead to the generation of undesired frequencies or harmonics. These non-linear effects can partly be influenced technologically by precise manufacturing, if they are e.g. B. be based on adhesion and edge inaccuracies. They can be electrical or mechanical in nature.

Use in practice

Impedance matching

AOW filters require an impedance adjustment in order to comply with the specifications (insertion loss, input and output reflection , transfer function). This is due to the fact that AOW filters as well as quartz and ceramic oscillators contain a capacitive component in the input and output impedance that has to be compensated for the frequency at which it is used.

The adaptation takes place with passive components such as coils and capacitors or also with stub lines . The real part of the input and output impedance of an AOW filter usually deviates from the impedance of the circuit in which it is operated - the input and output resistance are comparatively high.

Impedance matching can be achieved e.g. B. with a π filter, one capacitance of which is formed by the internal input or output capacitance of the filter. With π filters, the deviations in the real components of the impedances can also be matched to one another by varying the ratio of the capacitances.

Typical matching circuit of an AOW filter with coil and capacitor (internal capacitance forms π filter)

Another matching circuit is shown in the following figure; here a coil lying parallel to the input / output compensates part of the input / output capacitance of the filter.

Another matching circuit with coil and capacitor

Most often one finds matching circuits with a series connected coil and a parallel capacitor.

application areas

Pulse compression with an AOW filter. The distance between the "fingers" varies according to a certain law

The most common applications are intermediate frequency filters in radio receivers (cell phones, radiotelephones, television receivers, satellite receivers and video recorders). In television receivers (also in video recorders), the Nyquist transmission curve (a sideband filter with frequency-dependent attenuation) is implemented. In the case of voice radio, they can be used to achieve the narrow, as rectangular as possible, transmission curve (bandwidth, e.g. 9 kHz) between the receiving channels. With single sideband modulation ( SSB ), the sharp separation required for transmitters and receivers can be achieved with them.

Dispersive SAW elements were used in synthetic aperture radar devices to generate the chirp and for pulse compression. FIR filters are predominantly used as a software solution in modern devices .

literature

David Morgan: Surface Acoustic Wave Filters . 2nd Edition. Elsevier, Oxford, London, Burlington 2007, ISBN 978-0-12-372537-0 .
Franz Kubat: Theoretical and experimental investigations into the performance stability of surface wave filters . Cuvillier, Göttingen 2004, ISBN 3-86537-309-7 (dissertation; Faculty for Applied Sciences of the Albert-Ludwigs-University Freiburg im Breisgau).

Web links

Commons : SAW filters - collection of images, videos and audio files

Colin K. Campbell: Understanding Surface Acoustic Wave (SAW) Devices for Mobile and Wireless Applications and Design Techniques ( Memento from October 1, 2012 in the Internet Archive )