Vasil Postoyalko

Relationship between group delay and stored energy in microwave filters

In this paper, an expression for the time average stored energy (t.a.s.e.) in a passive lossless two-port is derived in terms of its scattering parameters. In particular, it is shown that the t.a.s.e. in a passive lossless reciprocal symmetrical or antimetrical two-port is proportional to the group delay. One implication of this result is that the t.a.s.e., which is linked to the power-handling capability in many passive filters used in practice, is proportional to the group delay of the filter. This rigorous derivation is based on a variational theorem, which has been used in the past to prove energy storage results for passive lossless one-ports and periodic two-ports.

Design of waveguide E-plane filters with all-metal inserts by equal ripple optimization

An optimization based approach to the design of E-plane filters is described. An optimization procedure based on Cohns equal ripple optimization is developed. This vector procedure has several advantages over the general purpose optimization routines previously applied to the design of F-plane filters. The problem of local minima does not arise. Optimization is carried out with respect to the Chebyshev (or minimax) criteria. Less frequency sampling and therefore less calculation of the electrical parameters of E-plane discontinuities is required. The design of a symmetrical F-plane filter is considered. Higher order mode interaction between E-plane discontinuities is not included in the design. For the design example considered this is shown not to be significant. A numerically efficient method, requiring only real scalar arithmetic, for calculating the insertion loss of a symmetrical cascade of lossless symmetrical 2-ports is employed. Measurements on a fabricated filter confirm the accuracy of the design procedure. >

Stopband performance improvement of rectangular waveguide filters using stepped-impedance resonators

Rectangular waveguide stepped-impedance resonators (SIRs) are analyzed and employed in the design of an X-band filter with center frequency f/sub 0/ = 10 GHz and a bandwidth of 100 MHz. An attenuation of 80 dB is held up to 23.1 GHz and, compared to standard uniform-impedance-resonator filters, a reduction in length of 55 % is achieved at the expense of an increased insertion loss from 0.6 to 1.5 dB. The second resonance of the fundamental TE/sub 10/ mode can be controlled by adjusting the length and impedance ratio of each resonator. A design procedure that takes into account step discontinuities is described and applied to the design of a number of SIR filters. Finally, the presented theory is supported with experimental results.

Stop-band improvement of rectangular waveguide filters using different width resonators: selection of resonator widths

Rectangular waveguide resonators having different widths can be mixed in order to improve the stopband performance of band-pass filters. Two effective procedures for the choice of the resonator widths are presented and implemented to realise X-band 6-cavity filters which hold 30 dB of attenuation over a frequency range 40% wider than a standard filter. Theoretical and experimental results are shown and commented on.

Optimum noise-source reflection-coefficient design with feedback amplifiers

The issue of designing a low-noise microwave feedback amplifier for a given optimum noise-source coefficient /spl Gamma//sub Sopt/ is addressed and a set of original formulas is presented. These expressions define a new procedure which does not rely on computer optimization in order to get the required noise performance of the low-noise amplifier stage. The technique permits the design of a circuit which is simultaneously noise and power matched at its input port without an input matching circuit. This method can be used to screen devices for an optimum noise performance and it provides the essential mathematical tool for designing the core of a feedback amplifier.

Prediction of peak internal fields in direct-coupled-cavity filters

In this paper, an explicit expression of the peak electric-field strength in the cavities of Chebyshev direct-coupled-cavity waveguide filters is derived. It is shown that the electric-field strength can be predicted from the analysis of the time-averaged stored energy in the lumped low-pass prototype, from which the cavity filter was derived. This simplifies the analysis and study of the power-handling capability of these types of filters considerably, as the stored energy in the prototype filters is easily computed. The analysis of the field distribution in the cavities of an example third-degree Chebyshev direct-coupled-cavity filter shows that the explicit expression for the peak electric-field strength derived in this paper agrees closely with results obtained from a TE/sub 10/ circuit model of the filter, from a full-wave electromagnetic solver and from measurements.

Analytical behavior of the noise resistance and the noise conductance for a network with parallel and series feedback

An analysis is presented of the changes of the noise parameters of a two-port network when noisy series and parallel feedback immittances are applied. Exact formulas for the noise parameters R/sub n/, g/sub n/, and /spl rho//sub n/ are given as functions of the feedback for a given network. It is proved that R/sub n/ always reaches a minimum when a reactive series feedback is considered. The same results are demonstrated for g/sub n/ since a duality principle is pointed out. The results are valid for a wide range of linear microwave two-port networks, either passive or active, and they are used to confirm the data from previously published work.

The Pospieszalski noise model and the imaginary part of the optimum noise source impedance of extrinsic or packaged FET's

The imaginary part X(S/sub opt/) of the optimum noise impedance for extrinsic or packaged devices is investigated. The analysis modifies the well-known Pospieszalski noise model by applying a series feedback to the source port. A simple expression for X(S/sub opt/) is developed and is verified for extrinsic and packaged devices with a decreasing level of accuracy. The results give further insights into the way the parasitic inductors L/sub g/ and L/sub s/ affect the noise performance of the transistor and can help to design low-noise amplifier with simultaneous signal and noise power match at the input port.

Filter topologies with minimum peak stored energy

In this paper the question of enhancing the power handling capability of resonator filters by choosing a suitable filter topology which maximally reduces the peak time averaged stored energy (t.a.s.e.) is addressed. A well known network transformation that alters the filter topology but leaves the 2-port parameter of the network unchanged is employed. After defining the cost function a stochastic search method is used to find the global minimum. Results presented include Butterworth filter topologies of degree 2 to 5.

Specifications for a linear network simultaneously noise and available-power matched

The authors address the problem of designing a linear lossy input matching network for low-noise amplifiers so that the source impedance can deliver its available power and correspond to the minimum noise figure of the driven stages. The differences between lossless and lossy networks are highlighted because matching circuits are usually considered to be lossless when designing an amplifier. After stating the assumptions, a solution to the problem of the minimum number of elements fulfilling the requirements is developed. The result explains why the standard distributed approach often fails to cope with minimum noise specifications when practical elements are considered.