Search

Equivalent schemes of an ideal operational amplifier and a difference operational amplifier are deduced. These schemes enable the analysis of networks containing operational amplifiers with the aid of a digital computer.

The effect of finite gain bandwidth product on a proposed network for the realization of all-pass functions with complex or real poles is examined. It is found that severe low-frequency limitations are imposed. A modification of the network is given in which this restriction may be overcome.

A single-tuned parametric amplifier was experimentally demonstrated at 18 GHz. A coaxial signal circuit and a radial idler cavity were used to obtain a large voltage-gain bandwidth product of 1440 MHz and a low noise temperature of 245/spl deg/K including the circulator loss contribution.

An analysis technique is proposed which takes into account the effect of finite DC gain and the gain-bandwidth product of operational amplifiers (OAs) on the response of switched-capacitor (SC) networks operated by a biphase or a multiphase clock. The general matrix equation of the network in the z-domain is split into two subsets: external charge conservation equations (CCEs), which are set up for the passive part of the network, and internal CCEs, based on equivalent OA representations. The technique is simple, versatile, applicable to SC networks of any topology and suitable for creating a computeraided analysis system. Examples are given for both to confirm the advantages of equivalent OA representations and to illustrate the technique's efficiency.

In this paper, the results are given of an investigation of first-generation integrated n-m.o.s. operational amplifiers. Calculator-aided simulation is used to study the performance and limitations of the stages and the overall amplifier. Quiescent conditions, differential-mode gain and common-mode gain are investigated. The transfer function of the overall amplifier is determined. The slew rate has been investigated, taking into consideration the complicated open-loop gain and the nonlinearity of the amplifier's different stages. Finally, an improvement in the circuit design of integrated n-m.o.s. operational amplifiers is suggested. This is especially suited to amplifiers fabricated by advanced technologies.

Ultra-wideband monolithic HEMT amplifiers have been designed and fabricated using MOCVD heterostructures. It has been shown that by taking into account the multiple reflections at the gate load, a very flat gain can be achieved and the sensitivity of the amplifier to the load resistances can be considerably reduced. By suitable design, the frequency bandwidth of the amplifier can be increased above the cutoff frequency of the active devices. In particular, it is shown that the cascode configuration reduces the coupling between gate and drain lines and increases the gain-bandwidth product, but can generate oscillations. To verify their design approach, the authors present the results obtained on 2-42 GHz monolithic amplifiers using 0.5 mu m HEMTs. >

This letter describes practical reflection-mode amplifiers utilising active-medium-propagation (a.m.p.) devices. A single-stage reflection-mode amplifier provided 24 dB gain and 350 MHz bandwidth at 15 GHz. A 2-stage reflection-mode amplifier provided 30 dB gain and 500 MHz bandwidth at 15 GHz.

We report very high-speed operation of InP/InGaAsP/InGaAs avalanche photodiodes with separate absorption, grading and multiplication regions (SAGM-APDs). For low multiplication values (M0≤7) the bandwidth of these APDs is relatively insensitive to the gain and is determined by the hole transit time and the RC time constant. In this gain region bandwidths as high as 5.5 GHz have been achieved. For higher multiplication values the frequency response exhibits a constant gain-bandwidth product. We have observed gain-bandwidth products as high as 40 GHz, the highest value reported to date for a device of this type.

The design, fabrication, and performance of a planar InP/InGaAs APD (avalanche photodiode) with high bandwidth for use in high-bit-rate lightwave systems are discussed. The InP multiplication region is highly doped to reduce the avalanche build-up region and increase the gain bandwidth product. This also lowers the transit time across the InP layer. An InGaAsP layer reduces hold trapping at the heterointerface. The thickness of the absorbing InGaAs layer is a compromise between reasonable quantum efficiency and short transit time. To further reduce transit-time effects, the InP buffer layer between the substrate and the InGaAs layer is highly doped to stop the depletion region at the edge of the InGaAs. This material structure is grown by VPE (vapor-phase epitaxy). Devices with gain-bandwidth products up to 70 GHz and bandwidths greater than 5 GHz at M=10 have been fabricated. Bandwidths at low gain of 7 GHz or greater, which can be explained solely by transit time effects have been measured. It is believed that a single InGaAsP layer in this material reduces hole trapping to negligible amounts at a heterointerface field of 6*10/sup 6/ V/m or greater. A strong dependence of gain bandwidth product on heterointerface field has been observed, even down to fields as low as 7*10/sup 6/ V/m. This can be explained by a small amount ( >

Reports the performance of the first high speed GaInAs/InP multi-quantum well avalanche photodiodes grown by atmospheric pressure MOCVD. The multi-quantum well avalanche region of the device consists of 50 periods of 150 AA wells and barriers forming the intrinsic region of a pin structure. Avalanche multiplication up to a factor of 25 has been measured at DC together with high-speed response giving a maximum measured RF gain of 16 and a gain-bandwidth product in excess of 25 GHz.

A biasing scheme for dynamic amplifiers is proposed and several dynamic amplifiers are described. Furthermore, advantages and limitations of dynamic amplifiers are briefly discussed and some experimental results are presented.

The exact mathematical solution of the amplifier finite gain and bandwidth problem in switched-capacitor filter networks is presented. The theory is incorporated into a computer analysis program and typical results for a filter circuit are included.

A procedure is developed for the design of amplifiers to meet specified gain and bandwidth figures. The number of required operational amplifier stages is minimised. A figure of merit is introduced to determine if a particular operational amplifier can be used to satisfy a given set of specifications.

The second-order effects of the gain-bandwidth product (g.b.w.) of operational amplifiers on the pole Q of certain active RC bandpass filters have been studied. These studies enable an accurate comparison of various configurations of active filters to determine their utility over a wide range of complex pole frequencies.

A new structure realising the voltage-scattering matrix of resistive n-port networks using resistors and operational amplifiers is proposed. The effect of a finite gain-bandwidth product of operational amplifiers is discussed.

A monolithic matched, two-stage wideband amplifier with an insertion gain of 26dB and a — 3dB bandwidth of 3-2GHz is reported. Optimally designed cascode circuits are used to enhance the gain-bandwidth product available per stage. The IC has been fabricated in a Iμ depletion GaAs MESFET technology.

A method for increasing the bandwidth of single-channel m.o.s. integrated inverters, with practically no decrease in low-frequency gain, is presented. A 1st-order analysis and computer simulation are included.

GaAlAs-GaAs heterojunction bipolar transistors (HBTs) with an abrupt emitter-base interface have been realised by molecular-beam epitaxy (MBE) on semi-insulating substrates. A gain-bandwidth product FT of 15 GHz has been measured for Ic = 20 mA and VCE = 8 V. These results are the best reported so far for HBTs and are very promising for high-speed logic.

An expression for the sensitivity of a general second-order RC-active filter section to the finite gain-bandwidth product of the active elements (assumed to be first-order operational amplifiers) is derived and used to classify a number of well known active filter configurations according to the values of certain parameters of the general section. The circuits with a good high-frequency performance are found to have a high sensitivity to the values of passive components.

The increase in bandwidth potential of a simple parametric amplifier obtained by adding active rather than passive filter elements to the signal circuit is discussed. Measured performance data of an amplifier constructed for the 4GHz satellite-communication band is given; by using this technique, a sevenfold increase in bandwidth at 20dB gain was obtained

It is shown that the basic action of the differential-input earthed-output type of operational amplifier may be given a flowgraph representation with a type of graph that is more general than those of Coates and Mason. Examples of embedded amplifiers are given to show the use of these graphs for the analysis of particular amplifier circuits.