Amplifier Input Research Articles

Hearing aid having switchable first and second order directional responses

A hearing aid apparatus is disclosed that employs both an omnidirectional microphone and at least one directional microphone of at least the first order. The electrical signals output from the directional microphone are supplied to an equalization amplifier which at least partially equalizes the amplitude of the low frequency electrical signal components of the electrical signal with the amplitude of the mid and high frequency electrical signal components of the electrical signals of the directional microphone. A switching circuit accepts the signals output from both the omnidirectional microphone and the directional microphone. The switching circuit connects the signal from the omnidirectional microphone to an input of a hearing aid amplifier when the switching circuit is in a first switching state, and connects the output of the equalization circuit to the hearing aid amplifier input when the switching circuit is in a second switching state. The switching circuit may be automatically switched in response to sensed ambient noise levels.

Operational amplifier immune to EMI with no baseband performance degradation

An operational amplifier (opamp) input stage achieving a high degree of immunity against electromagnetic interference (EMI) is presented. Unlike previously proposed topologies, the novel circuit does not show any significant penalty in baseband operation. The effectiveness of the proposed solution and its robustness to mismatch and process variations is tested by simulations performed with reference to the AMS C35 0.35 µm CMOS technology.

Linear amplification by time-multiplexed spectrum

A new waveform processing technique for reducing intermodulation distortion (IMD) generated by an amplitude-modulated signal is presented. The technique eliminates amplifier-generated IMD by time-multiplexing portions of the signal spectrum prior to amplification. The input signal is described as a sum of sinusoidal signals. The peak-to-average ratio of the time-multiplexed amplifier input is lower than that of its non-multiplexed counterpart, but it contains spectral aliases at multiples of the switching frequency. Thus, the reduction in distortion of the desired amplified signal is achieved at the expense of momentarily widening the signal bandwidth. Following amplification, the desired signal is recovered using a bandpass filter which spreads the signal in time and narrows its bandwidth. An analytical expression for distortion reduction when amplifying two multiplexed carriers is developed along with measurements verifying the theory. Distortion reduction is demonstrated experimentally at 3.6 GHz for 2–20 multiplexed carriers, and linear recovery is demonstrated for four multiplexed carriers. The technique reduces third-order IMD by 8–22 dB in experimental measurements of three different amplifiers. The presented technique has the unique property of improving linearity without requiring feedback, feedforward cancellation or calibration.

On the Benefits of Optical Gain-Clamped Amplification in Optical Burst Switching Networks

This paper investigates the performance of an all-optical method for amplification gain control to be applied in the next generation of optical networks. An erbium-doped fiber amplifier is implemented in a simple and passive all-optical configuration known as optical gain-clamped optical amplifier (OA). The paper investigates the dynamic performance of the OA and discusses the interplay of amplifier dynamics with different traffic statistics. The investigation concerns exhaustive characterization of bit error rate performances under typical optical burst switching (OBS) traffic as well as special case of sudden power variation at the amplifier input. All obtained results show a reduction in the amplifier output power overshot compared to the case where the same OA operates without any gain stabilization technique. As an example, in the typical OBS traffic scenario, a reduction of 3 dB is observed.

Mixed-Signal Architectures for High-Efficiency and Low-Distortion Digital Audio Processing and Power Amplification

The paper addresses the algorithmic and architectural design of digital input power audio amplifiers. A modelling platform, based on a meet-in-the-middle approach between top-down and bottom-up design strategies, allows a fast but still accurate exploration of the mixed-signal design space. Different amplifier architectures are configured and compared to find optimal trade-offs among different cost-functions: low distortion, high efficiency, low circuit complexity and low sensitivity to parameter changes. A novel amplifier architecture is derived; its prototype implements digital processing IP macrocells (oversampler, interpolating filter, PWM cross-point deriver, noise shaper, multilevel PWM modulator, dead time compensator) on a single low-complexity FPGA while off-chip components are used only for the power output stage (LC filter and power MOS bridge); no heatsink is required. The resulting digital input amplifier features a power efficiency higher than 90% and a total harmonic distortion down to 0.13% at power levels of tens of Watts. Discussions towards the full-silicon integration of the mixed-signal amplifier in embedded devices, using BCD technology and targeting power levels of few Watts, are also reported.

Integration Challenges in Standard CMOS with Multiple Gate Oxide Thicknesses

Following the prototype release of an automotive, dual gate oxide, and high voltage process, yield problems were observed on an intermittent basis. An investigation was made into a temperature sense circuit and a differential input of a unity gain operational amplifier used for I/O circuitry. The problem was isolated to a bimodal distribution of threshold voltage values for standard gate oxide thickness transistor test structures. When all such structures on the wafer were electrically probed, their distribution of failures was found to be mostly random. Curiously, this mismatch was consistent with a small increase in gate oxide thickness but not observed in the single gate technology. Microscope inspection following etchback of oxide from standard gate regions provided evidence of water spotting and damage to active areas. Process changes were implemented to minimize this damage. Subsequent lots yielded well and matching problems were eliminated.

A High-Gain Acquisition System With Very Large Input Range

A signal acquisition system is presented for the recording of small signals in the presence of low-frequency interference. In a conventional design, high-pass filters precede the input stage to remove the impeding signals. However, filters limit the recording bandwidth, can lead to long settling time after disturbance, and potentially valuable information about the interfering signal is removed. Also, the required large components are impractical for a fully integrated solution. In this paper, a flexible, low-cost, filterless system is proposed and demonstrated, which extends the available input range to accommodate interference instead of rejecting it. An integrated system using digitally controlled floating batteries to compensate for interference at the amplifier input is presented. A digital postprocessing algorithm suitable for integration is proposed and evaluated. Measured results of the acquisition front-end realized in 0.35 mum CMOS technology confirm a more than sevenfold extension of the original input range and the usable dynamic range. The system is further verified using electrocardiography as a typical biomedical application.

Spectral broadening of femtosecond pulses in an nonlinear optical fiber amplifier

A promising method for the generation of a supercontinuum with a high spectral power density based on the spectral broadening of ultrashort pulses in a fiber amplifier is considered. The advantage of the method, as compared to the conventional way of the supercontinuum generation in a microstructure fiber, is a lower pulse spectral broadening rate, which allows one to achieve higher SC spectral power densities. The initial stage of the supercontinuum generation in an ytterbium fiber amplifier (a fiber core diameter of 7 µm) with side pumping from an array of laser diodes with a total power of 8 W at a wavelength of 976 nm is experimentally studied. Yb:KYW laser pulses with a duration of 250 fs, a central wavelength of 1046 nm, and an average power of 150 mW have been supplied to the input of the amplifier. In this case spectrally broadened radiation with an average spectral power density of higher than 65 mW/nm and a spectrum width of 50 nm has been obtained.

An Integrated 800-MHz Coupled-Resonator Tunable Bandpass Filter in Silver With a Constant Bandwidth

This paper presents a fully integrated tunable lumped filter on silicon using a low-temperature silver micromachining process. A prototype 836-MHz bandpass filter with a 3-dB bandwidth of 5.9% and insertion loss of 4.8 dB is demonstrated in a second-order coupled-resonator configuration. Continuous tuning of 50 MHz is achieved by electrostatically actuating the lateral air-gap capacitors of the filter. To control the bandwidth while tuning the center frequency, reconfigurable termination impedance is proposed. As a proof of concept, a low-noise amplifier with tunable input impedance is designed to interface with the bandpass filter. The tunable impedance is realized at the input of the low-noise amplifier using a shunt positive metal-oxide-semiconductor transistor. The fabrication, design, and measurement results of the filter are detailed, and future research directions to improve the performance of such filters are discussed.

A digital X-ray imaging system based on silicon strip detectors working in edge-on configuration

We present the energy resolution and imaging performance of a digital X-ray imaging system based on a 512-strip silicon strip detector (SSD) working in the edge-on configuration. The SSDs tested in the system are 300 μm thick with 1 or 2-cm-long strips and 100 μm pitch. To ensure a very small dead area of the SSD working in edge-on configuration, the detector is cut perpendicular to the strips at a distance of only 20 μm from the end of the strips. The 512-strip silicon detector is read out by eight 64-channel integrated circuits called DEDIX [Grybos et al., IEEE Trans. Nucl. Sci. NS-54 (2007) 1207]. The DEDIX IC operates in a single photon counting mode with two independent amplitude discriminators per channel. The readout electronic channel connected to a detector with effective input capacitance of about 2 pF has an average equivalent noise charge (ENC) of about 163 el. rms and is able to count 1 Mcps of average rate of input pulses. The system consisting of 512 channels has an excellent channel-to-channel uniformity—the effective threshold spread calculated to the charge-sensitive amplifier inputs is 12 el. rms (at one sigma level). With this system a few test images of a phantom have been taken in the 10–30 keV energy range.

Analysis of nonlinear effects and calculation of EMC characteristics of SHF amplifier based on heterojunction bipolar transistor

Analysis of nonlinear properties of amplifier, based on heterojunction bipolar transistor, under the influence of intensive interferences at the amplifier input was carried out with Walter’s series method. It is shown that nonlinearity level and character depend on mutual compensation of harmonics, which are generated into base and collector circuits, and it is also dependent on correlation between Taylor expansion coefficients of nonlinear elements of equivalent circuit near operating point and transistor input impedance.

RG64—High Count Rate Low Noise Multichannel ASIC With Energy Window Selection and Continuous Readout Mode

We report on the performance of a low noise and high count rate readout ASIC with binary architecture and energy window selection for X-ray imaging applications using semiconductor detectors. The ASIC called RG64 is designed in 0.35 mum CMOS process and its total area is 3900times5000 mum2. The core of RG64 consists of 64 readout channels. Each channel is built of a charge sensitive amplifier with a second order shaper of peaking time 75 ns, two independent discriminators with an 8-bit offset correction circuit and two independent 20-bit counters with RAM memory buffers. The ENC of the circuit reaches the value of about 126 el. rms with 1 pF input load and 5 mW power consumption per single channel. The mean gain in the multichannel ASIC is about 50 muV/el., with the dispersion from channel to channel of 0.9% (on one sigma level). The deviation of the effective threshold voltage spread for given energy can be reduced to less than 7 el. rms (calculated to the charge sensitive amplifier input). High count rate measurements have been performed up to 2 Mcps of average rate of input pulses, both for AC and DC coupled silicon strip detectors with X-ray photons of energy 8.04 keV. The RG64 can operate both in the continuous readout mode and in the readout mode separate from exposure.

Microstrip superconducting quantum interference device radio-frequency amplifier: Effects of negative feedback on input impedance

We present the results of measurements of the scattering parameters of microstrip amplifiers (MSAs) based on the dc superconducting quantum interference device. The amplifier input impedance is poorly matched to typical transmission line impedances, resulting in high input return loss around −2 dB. We show that negative feedback can lower the MSA input impedance to achieve a robust match to 50 Ω. In the presence of capacitive and inductive feedback, the input return loss of the MSA can be reduced below −10 dB, opening the door to the practical use of the MSA for a variety of demanding applications.

Magnetic markers detection using PCB fluxgate array

We used an array of race-track fluxgate sensors, manufactured with printed circuit board (PCB) technology, forming a sensor head for detection of ferromagnetic and paramagnetic markers. The sensors were arranged perpendicularly to the measuring plane and we measured the difference of their output, giving us the horizontal gradient of normal component of the measured field. Due to the close match of the sensor’s parameters, subtraction of the fluxgate output signals could be done directly at the input of a lock-in amplifier, increasing the signal-to-noise ratio for small gradients. When moving the sensor head, we were able to map field gradients smaller than 6 nT/mm, which was verified while measuring the magnetic markers on a dollar bill, while suppressing the background field by a factor of 5. In a line-scanning mode, we scanned a marker formed by a 0.2 mm diameter Permalloy wire in a distance of up to 10 mm. With the help of perpendicular ac excitation at 30 Hz, we were able to detect a 0.1 ml Endorem iron-oxide superparamagnetic marker at 2 mm; volume of 0.6 ml was detectable at 10 mm.

Utilidad de la monitorización electromiográfica durante la estimulación eléctrica cortical intraoperatoria en áreas motoras

Surgery for resection of supratentorial lesions near the motor areas and/or internal capsule carries an associated risk of damage to cortical or subcortical motor pathways. Intraoperative brain mapping, using direct cortical electric stimulation, is a surgical adjunct used during lesionectomy in functional cortex. The technique of observing movements elicited by electrical stimulation has proved useful for intraoperative localization of motor pathways but it is difficult observe the entire body at once. Thus, a small movement may be missed while attention is focused on another site. To determinate the usefulness of electromyographical monitoring in locating motor pathways during neurosurgery in functional cortex. We recorded multichannel electromyography activity during cortical and subcortical electric stimulation in a consecutive series of 40 patients undergoing craniotomies for lesionectomy near motor cortex or subcortical motor tracts. Because the aim of the electromyography recordings was to sample as many muscles as possible, the two electrodes connected to each different amplifier input were placed in different muscles in the same region of the body. No motor responses were detected by either electromyographical monitoring or visual inspection in 20% patients. Electromyographical activity was the only indication of motor response in 17%. Motor responses were detected simultaneously by either electromyographical monitoring and visual inspection in 76%. Electromyography monitoring enhances the ability to detect the location of primary motor cortex and subcortical pathways with electrical stimulation during neurosurgery.

Micro-scale thermal tissue gripper

During eye surgery translocating a graft of retinal tissue without damaging the vulnerable top layer is highly problematic using the currently available instruments. This study presents indirect Heat Induced Attachment and Detachment (HIAD) as a new concept for single side attachment of tissue to, and detachment from a heated metal wire.A small-scale prototype was built, having a 50 μm molybdenum wire that could be heated via an electric current. Tests (n = 60) were performed on submerged chicken meat to study the attachment and detachment properties of the prototype at different electric signal lengths. A 9V signal was applied to the prototype, with amplifier input signal lengths varying from 0.6–1.5 ms. Voltages and currents were sampled at 250 kHz to obtain energies.Both attachment and detachment occurred with 98% and 90% certainty, respectively, supplying 41 and 110 mJ of energy to the prototype in 0.7 and 1.5 ms. The attachment strength was estimated at 0.2 mN. Visible damage appeared to be approximately 0.005 mm2.The concept of indirect heating of the instrument-tissue interface proved to be effective as the prototype could induce attachment and detachment of tissue. Indirect HIAD may be applicable in many different surgical applications.

0.9 V rail-to-rail constant gm CMOS amplifier input stage

Presented is a 0.9 V rail-to-rail constant gm CMOS amplifier input stage consisting of complementary differential pairs and a gm control circuit. The gm control circuit eliminates the gm dead zone, which occurs in the conventional rail-to-rail amplifier with ultra-low supply voltages. The proposed amplifier input stage has a constant gm that varies by ±2.3% for rail-to-rail input common-mode levels. To verify the proposed amplifier design, an experimental prototype operational amplifier is also implemented using 0.35 µm standard CMOS technology.

Frequency reconfigurable RF circuits using photoconducting switches

Designs for a frequency switchable dual-band branch-line coupler and a reconfigurable S-band power amplifier input matching network with photoconducting switches are presented. Frequency switching is achieved by increasing the power of the laser applied to the highly resistive silicon wafer and changing the properties of silicon under optical illumination. The advantages of this approach are high-speed switching, electromagnetic transparency (no interference), and thermal and electrical isolation between the device and the control circuit. A branch-line coupler frequency shift of 35% and 10% has been achieved from all switches off to all switches on in lower (900 MHz) and upper (1800 MHz) frequency bands, respectively. Frequency switchable class AB power amplifier with silicon switch in the input matching circuit has obtained the frequency tuning range of 2.5–3.5 GHz with no significant loss in efficiency and linearity. © 2009 Wiley Periodicals, Inc. Int J RF and Microwave CAE, 2010.

Reducing the EMI Susceptibility of a Kuijk Bandgap

In this paper, the susceptibility of a Kuijk bandgap voltage reference to electromagnetic interferences (EMIs) superimposed to the power supply is investigated. A model of the bandgap circuit is derived from experimental tests consisting of scattering parameters and susceptibility measurements on a test chip. The model is able to correctly predict the susceptibility of the circuit by means of SPICE transient simulations. The simulations identify the fundamental stray components responsible for EMI coupling that are usually not taken into account in postlayout analyses. From the EMI point of view, the rectification phenomenon of bipolar transistors, used in the bandgap cell, and the operational amplifier (op-amp) input distortion are shown to cause the voltage reference performance degradation. In particular, for the first time, the importance of the bandgap cell is pointed out, suggesting that the measures to reduce the susceptibility of the op-amp may not be sufficient to guarantee the bandgap immunity. Moreover, the analyses of main parasitic paths (from the power supply net to the more sensitive nodes) demonstrate the relevance of paths that are not commonly considered during the design phase and that may lead to an immunity degradation. Possible solutions to reduce the bandgap susceptibility are also explored, suggesting design criteria, filtering techniques, and layout variations. A second test chip is designed and manufactured to validate the suggested improvements and verify the critical role of the bandgap cell. Finally, a simplified theoretical analysis, which allows a fast bandgap susceptibility evaluation, is presented. This tool is used to point out the importance of the rectification phenomenon compared to the op-amp differential pair distortion.

X-Calibration: A Technique for Combating Excessive Bitline Leakage Current in Nanometer SRAM Designs

In an SRAM circuit, the leakage currents on the bit lines are getting increasingly prominent with the dwindling of transistors' threshold voltages as the technology scales down to 90 nm and beyond. Excessive bit-line leakage current results in slower read operations or even functional failure. In this paper, we present a new technique, called X-calibration, to combat this phenomenon. Unlike the previous method that attempts to compensate the leakage current directly, this scheme first transforms the bit-line leakage current into an equilibrium offset voltage across the bit-line pair, and then simple circuitry is utilized to cancel this offset accurately at the input of the sense amplifier so that the sensing is not affected by the bit-line leakage. SPICE simulation of a 1 Kbit SRAM macro shows that this X-calibration scheme can handle 83% higher bit-line leakage current than the previous bit-line leakage compensation scheme. Measurement results of the test chip show that the SRAM macro adopting X-calibration scheme can cope with up to 320 muA bit-line leakage current.