Low-loss slanted SAW filters with low shape factors

Chvets, V. B.; Ivanov, P. G.; Makarov, V. M.; Orlov, V. S.

doi:10.1109/ultsym.1999.849354

Cited by 13 publications

(1 citation statement)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All the SFIT structures are characterized by finger arrangements like a divergent bundle. Later, in particular in the 90s, SFITs became important for wide band filters with flat passband, steep skirts, and good stopband rejection [2]- [11]. Saw and Campbell [9] as well as Solie [10] suggested to combine tapered electrode structures and single-phase unidirectional transducers (SPUDT) cells.…”

Section: Introductionmentioning

confidence: 99%

SAW filters including one-focus slanted finger interdigital transducers

Martin

Steiner

2003

IEEE Trans. Ultrason., Ferroelect., Freq. Contr.

View full text Add to dashboard Cite

The present paper describes a new filter type, slanted finger interdigital transducer (SFIT) that allows fast analysis as a precondition for fast optimization. Therefore, filter structure and analysis are especially suitable for optimizing unidirectional SFIT (USFIT) filters. For analysis, a SFIT filter is usually divided into many narrow channels. Every channel is considered to be a subfilter and is analyzed separately. Therefore, the total filter analysis is very time consuming. For reducing computing time, we suggest a one-focus structure. In contrast to conventional SFIT filters, not only the finger and gap widths but also the spaces between transducers of different channels differ by one and the same scaling factor. As a consequence, all prolonged finger edges of both transducers intersect in one point called focus. As a result, the parameters of all subfilters can be calculated from the parameters of only one subfilter by simple frequency scaling. Consequently, the total time for analysis is essentially reduced. However, one-focus SFIT filters with continuous finger edges show a deep minimum within the passband. This problem can be overcome by using stepped one-focus structures.

show abstract