Showing papers on "Feedback loop published in 1992"

Multibit Sigma - Delta A/D converter incorporating a novel class of dynamic element matching techniques

Abstract: A dynamic element matching technique is applied to multibit sigma-delta modulators. The approach translates the harmonic distortion components of a nonideal digital-to-analog converter (DAC) in the feedback loop of a sigma-delta modulator to high-frequency components, which can then be filtered out by the decimation filter. Computer simulations have confirmed that with this approach a third-order sigma-delta modulator employing a 3-bit forward ADC, a 3-bit feedback DAC with a random mismatch of 0.1% can achieve a 104-dB (17+bit) dynamic range and a harmonic distortion below 100 dB, with an oversampling ratio of 64. The technique does not generate any tone in the passband. >

Physical origins of input filter oscillations in current programmed converters

Abstract: Addition of an input filter to a current-programmed converter can cause the controller to oscillate. Two instability mechanisms can typically occur: (1) the current programmed controller effective current feedback loop may become unstable, or (2) the controller effective input voltage feedforward loop, which becomes a positive feedback loop when an input filter is added, may oscillate. Three design criteria for preventing oscillations are derived and interpreted. When all three criteria are well satisfied, then the output voltage regulation loop gain is unchanged. Hence, input filters of current programmed converters can be designed in essentially the same manner as for duty-ratio programmed converters. Results are summarized in tabular form for the basic buck, boost, and buck-boost converters. Experimental measurements for a buck converter with different input filters support the theoretical predictions. >

A comparison of two approaches to feedback circuit analysis

Abstract: Two approaches for analyzing single-loop feedback circuits are compared and contrasted. One approach is based on the return-radio concept, and the other is based on two-port analysis. A frequent error in many popular texts-interchanging the computation of return ratio for a dependent source and loop gain of the idealized feedback network-is discussed. Assumptions commonly made in many texts when presenting two-port feedback analysis are examined for validity. Examples are given to highlight the differences between the two approaches. >

Common-mode feedback in electrical impedance tomography

Abstract: When a current is injected into a body, in addition to the voltage profile developed on the surface, a common-mode voltage (CMV) which produces errors in the measurement also appears. The great accuracy needed to reconstruct images in electrical impedance tomography (EIT) requires the use of differential amplifiers with a high common-mode rejection ratio (CMRR) to avoid this error. Nevertheless, the effective CMRR is lower than the differential amplifier ratio due to mismatches in the electrode impedances and other circuits in the measurement channel. The use of common-mode feedback (CMFB) is an alternative to reducing the error produced by the CMV. The stability of the feedback loop is analysed for a broadband system. Simulation and experimental results show that it is possible to obtain an improvement of 40 dB in the measurements at frequencies of up to 10 kHz.

The combined analogue locked loop universal modulator (CALLUM)

Abstract: The CALLUM modulator is proposed. By embodying the LINC component separation concept within a closed-loop feedback structure, it permits the realization of a highly efficient linear transmitter using practical components and requiring minimal calibration. The CALLUM modulator is unique in that it achieves frequency translation and component separation simultaneously using a pair of voltage controlled oscillators within the feedback loop. Several prototype linear transmitters have been built using this approach which have demonstrated the robustness of the technique. >

Remarks on a neural network controller which uses an auto-tuning method for nonlinear functions

Abstract: When the neural network is applied to a dynamic system controller, the stability of the neural network controller must be guaranteed. Stability is related to the optimum sigmoid function shape. An autotuning method for the optimum sigmoid function is proposed. The automating method is applied to a learning type direct controller in order to confirm its characteristics. The neural network of this controller has three layers with no inner feedback loop and no direct connection from the input layer to the output layer. Both the input and the hidden layers have four neurons and the output layer has one neuron. Both the hidden and output layers have a sigmoid function to provide the nonlinear mapping capability. The autotuning method uses the steepest descent method in order to apply it to servo control systems. Simulation results using the learning type direct controller confirm that the autotuning method is useful in combination with weight tuning for dynamic systems. >

Automatic level control system and method

Abstract: An automatic level control system and method provide continuous proportional control of an inlet valve when the level of material in a storage container is between low and high limits. The system includes a main feedback loop and a nested feedback loop to insure proper control of the inlet valve through which material is added to the storage container. The main feedback loop provides a control signal indicating the desired setting of the inlet valve, and the nested feedback loop provides a control signal indicating the actual position of the inlet valve so that the two control signals can be compared for use in accurately moving the inlet valve to the desired position. The main feedback loop includes a computer into which different density values can be input for accurately determining the amount of material in the container as indicated by a weight sensor.

Variable Structure Digital Repetitive Controller

Abstract: Repetitive controllers as a new type of servo mechanism have been successfully applied to mechanical systems such as robot and disk drive track following. Many research in this field have been devoted to stability and robustness considerations of adding a repetitive signal generator in the feedback loop. In this paper, a variable-structure digital repetitive controller, which can adapt a slow varying period, is proposed. For accurately identified plant, it is shown that computations involved in identifying a period can be substantially reduced. Furthermore, an FIR filter is added to the period identifiers in order to cope with model uncertainties. It is shown that for small perturbations, unbiased estimate of period can be achieved by adding a low-order FIR filter.

MR preamplifier having feedback loops for minimizing differential low frequency components and regulating common mode low frequency components of the preamplifier output signal

Abstract: A preamplifier circuit amplifies signals from a magneto-resistive element. The preamplifier circuit utilizes a voltage regulator to regulate one terminal of a magneto-resistive element at a predetermined voltage. A first feedback loop is used to minimize a differential DC component and a differential low frequency component of the differential output signal. A second feedback loop is utilized to cause a common mode component of the differential output signal to track an externally supplied reference voltage. The preamplifier of the present invention protects the magneto-resistive elements from short circuits of the disc medium, is powered by a single-ended power supply, and is capable of being coupled with other circuitry without physical coupling capacitors.

Run/standby control with switched mode power supply

Abstract: A television receiver has a switched mode power supply for regulating output voltages including the B+ output to a flyback transformer, by generation of output pulses to a power transformer. A controller for the power supply is coupled in two feedback loops, one responsive to a pulse width modulator coupled to the flyback transformer and the other comparing the output voltage with an internal reference for free running operation. The first feedback loop takes precedence and is active in the run mode of the television receiver. In the standby mode when horizontal rate pulses are absent, the second feedback loop takes over. The second feedback loop has a different reference level than the first, such that when switching from the run mode to the standby mode a transition interval occurs in which no pulses are output by the controller, thereby causing the B+ voltage to the flyback transformer to fall. During this transition interval, horizontal scanning signals continue at falling amplitude, and a signal is input to the kinescope drivers, thereby collapsing the picture and draining the ultor voltage on the screen anode. The reoccurrence of controller pulses when the output voltage of the power transformer has fallen to the new lower reference level is used to positively switch operation of the receiver at the conclusion of the transition interval.

Method and apparatus for decision feedback equalization with reduced convergence time

Abstract: A method and apparatus for performing blind mode decision feedback equalization with reduced convergence time. In operation, the decision feedback equalizer (DFE) of the invention receives a transmitted signal x(n) resulting from transmission of an input signal s(n) through a transmission channel and a receive filter. The apparatus includes a DFE feedback loop having an adaptive filter designed to converge rapidly and iteratively during blind mode decision feedback equalization to a best approximation A'(z) of the unknown, combined, transmission channel and receive filter transfer function A(z). In a preferred embodiment, the apparatus includes DFE circuitry for implementing a Widrow-Hoff least-mean-squared adaptation process, including circuitry for generating an input signal replica s'(n) and an error signal e(n) during each iteration of the adaptation process. In this embodiment, the adaptive filter adaptively generates N replica coefficients a' j (n), where j=1, 2, . . . , N, which define approximation A'(z), and a replica coefficient a' O (n) which is employed to update the error signal e(n).

Approximate Identification in view of LQG Feedback Design

Abstract: Due to the modelling error a model-based controller generally works better with the model than with the modelled plant. This difference between the performances can be made small by selecting a model that is accurate at the closed loop relevant frequencies. In this paper it is shown that an iterative approach of identification and control design can lead to a model that is much better suited for feedback design than a model resulting from a plain open loop identification. In this iteration each identification is performed such that a certain closed loop criterion function is minimised. Each control design step employs the latest identified model to construct an LQG compensator. The performance requirements are gradually increased during the iteration.

Application of H-infinity theory to robot manipulator control

Abstract: The feasibility of H-infinity control theory for robot manipulator control is investigated. Given a model of manipulator dynamics, a control methodology is developed which utilizes an inner loop linearizing decoupling feedback control and an H-infinity outer loop compensator. A methodology for handling model simplifications in the linearizing feedback loop is also developed. An example control design is presented along with simulation results. Robust tracking of trajectories is shown, given bounds on model parameters. >

Ultrasonic position sensor for measuring movement of an object

Abstract: A position sensor for measuring the distance from a first position to a second position. The sensor may include an elastic wave generator, an acoustical path over which the elastic wave passes, and a transducer to which the wave passes. A measurement loop includes the acoustical path, and has electrical and optical elements. The electrical elements may compare the phases of the transducer signals from which the change in position between the first and second positions are determined. Alternatively, the measurement loop may comprise a feedback loop to produce oscillations and an output frequency. A frequency counter is included in the feedback loop to measure frequency.

Parallel waveform relaxation of circuits with global feedback loops

Abstract: Feedback loops often severely degrade the performance of waveform relaxation techniques in solving large circuit analysis problems. Several new approaches have been studied to provide greater parallelism and faster convergence for such circuits. WRV256, an experimental waveform-relaxation-based parallel circuit simulator for the Victor family of distributed memory parallel machines, was used to study performance tradeoffs of partitioning and scheduling algorithms for circuits containing global feedback loops. This investigation included circuits ranging from less than 300 to over 93000 transistors. Several of the circuits were extracted directly from a 16 Mb DRAM design. >

Stabilization of a rotating rigid body by the energy momentum method

Abstract: The authors discuss the stabilization by feedback control of a rigid body uniformly rotating about an arbitrary axis. They consider a class of feedback laws that depend on a parameter matrix W, which is nonsingular and symmetric. It is shown that under this class of feedback laws, the closed loop system is Hamiltonian and represents the motion of a generalized rigid body. For such systems the energy-momentum method enables determination of feedback gains which ensure the stability for the closed loop system. Several examples are used to demonstrate the methodology. >

A database driven fast feedback system for the Stanford linear collider

Abstract: A new feedback system has been developed that stabilizes the SLC beams at many locations. The feedback loops are designed to sample and correct at the repetition rate of the accelerator. Each loop can be distributed across several INTEL 80386 microprocessors that control the SLC hardware. A new communications system, KISNET, has been developed to pass data between the microprocessors at this rate. The software is written using the state space formalism of digital control theory and is database driven. This allows a new feedback loop to be implemented by setting up the on-line database and perhaps installing a communications link. Eighteen such loops have now been implemented and this has measurably improved the performance of the accelerator.

High bandwidth information channel with optocoupling isolation

Abstract: An accurate direct current to 1 MHz (or more) analog information channel with 7.5 kV input to output isolation is provided using a commercially available opto-electronic isolator. The relatively slow response, (limited bandwidth) of the LED portion of the opto-electronic isolator is overcome by providing a frequency compensated negative feedback loop around the servo photodetector in the opto-electronic isolator. The feedback loop includes proper frequency compensation for the phase lag properties of both the LED and the servo photodetector, thus extending the bandwidth well beyond the 200 KHz typical of such opto-electronic isolators.

A review of quantitative feedback theory as a robust control system design technique

Abstract: Performance robustness has been an important issue in control and it has been increasingly recognised as an area of significance in many design applications. This issue has particular importance among the engineering community because it offers a solution to control system design problems for plant and processes that cannot be described by a single linear time-invariant model. The aim of this paper is to describe one of the methods available for robust control system design called Quantitative Feedback Theory (QFT) and discuss its application to engineering problems. QFT is a powerful design technique based on multi degree-of-freedom use of feedback. The basic building block of a multi-degree-of-freedom system is a two-degree-of-freedom system in which two controllers are present, one inside the feedback loop, used to reduce closed-loop uncertainty, and the other prior to the loop, used to shape the input in order to achieve the required output. To show the power of the method, the application to an aero ...

Voltage loop design for a low-ripple fast-response AC/DC switchmode magnet power supply

Abstract: Voltage feedback loops for AC/DC switch-mode magnet power supplies are designed and analyzed. In order to achieve low-ripple fast-response voltage loop performance, proper structures of feedback controllers and output filters are identified. In addition, feedback loop stability is analyzed, and the safe operation region of the system is determined, taking into account the nonlinearity of the pulse-width modulator. Results from an experimental breadboard are provided for verification. >

Vector modulation wave signal generator

Abstract: PURPOSE:To calibrate a level of a modulation wave signal subject to vector modulation attended with amplitude modulation stably with high accuracy and to output the result. CONSTITUTION:A modulation section 1 and a loop forming means 3 being major sections are switched into the calibration state for the calibration with a control means (switch means) 5 or the test state to output a modulation wave signal at a prescribed level based on the calibration. The modulation section outputs a carrier signal of non-modulation and outputs a modulated modulation wave signal in the test state. The loop forming means holds and gives a same control voltage as that for forming a feedback loop in the calibration state to a control terminal of the level variable circuit 2 when the feedback loop is opened in the test state.

Dual feedback path linearization of current modulated laser with dual amplifier circuit

Abstract: A laser driver mechanism for image reproduction maintains linear operation over a wide frequency an dynamic range of amplitude-modulated input signals by means of a first feedback loop and a second feedback loop. The second feedback loop contains a first amplifier coupled to receive the input signals and an output coupled to a second amplifier that drives an injection laser diode. A photodetector is coupled downstream of and external to a sealed unit in which the injection laser diode emitter, an associated photodetector of the first feedback loop and beam extraction optics are housed. The photodetector of the second loop has an output coupled to the second amplifier, so that the second feedback loop serves to compensate for non-linearities in the laser driver including those of the first feedback loop. Because the second photodetector is supported externally of the sealed unit, the feedback path from the second photodetector effectively forms an outer loop feedback path relative to the first feedback path, so that the ambient operating conditions of the second photodetector are not affected by the operation of the laser. The sealed unit is mounted in a module that effectively equalizes the cross-sectional spatial distribution of optical energy within the light beam emitted by the injection laser diode and ensures that the spatial cross-section of the image spot seen by the photodetector is effectively constant regardless of laser output beam intensity. For precision scanning of the recording medium, the laser output beam is scanned by a linearly swept galvanometer.

Complexity and Feedback

Abstract: In many scientific fields an increasing attention is being devoted to problems connected with complexity, as testified, for example, by the recent book by Nicolis and Prigogine [1].

Three-loop feedback control of fault-tolerant power supplies in IBM Enterprise System/9000 processors

Abstract: In an Enterprise System/9000™ (ES/9000™) processor, a fault-tolerant power system composed of multiple power supplies connected in parallel provides thousands of amperes of current to low-voltage (1–2 V) logic circuit boards, monitors the voltage at each board, and immediately responds to compensate for failure of a supply. If a supply fails, the very fast closed-loop response redistributes the current uniformity among the remaining supplies and allows the normal functioning of the processor logic to continue uninterrupted. This rapid response is not obtained from a conventional two-loop (current-mode) feedback power supply because the loop bandwidth is restricted by a resonance that develops in the power distribution. A third feedback loop that is added to each supply controls this power distribution resonance and makes possible the wide loop bandwidth necessary to achieve the required power system control. Analysis is presented of a three-loop control system, and a simulation of its application to a typical ES/9000 power system is described.

Method of creating synthesized rail profile.

Abstract: A sensed force signal is averaged over a plurality of trips under differing load conditions and the average is pre-stored for retrieval at various vertical points in the hoistway and may be used, in combination with a feedback loop, to predict horizontal forces about to act on the car 30.

Feasibility study of current mode differential amplifier design

Abstract: A new class of current mode amplifiers, designed to operate in a true current-mode environment, is introduced. They utilize the FCS (floating current source), a novel current-mode building block, in various configurations. Two current-mode differential amplifiers of this type are described. The first one employs two translinear gain stages. The CM (common mode) response is reduced by a negative feedback loop responding to the CM only. The translinear gain can be adjusted by varying the value of two current source pairs in unison. The second is a differential feedback amplifier, employing two independent feedback loops which apply parallel feedback to the input virtual ground, one for the differential and another for the common mode. Its input admittance and output impedance are both increased significantly by the feedback applied. >

Linear amplifier having path time delay compensation

Abstract: A linear amplifier (100) comprising a feedback loop (195) is stabilized by determining phase error between an error signal (111) and a feedback signal and shifting the phase of a forward signal (103) accordingly.

An Improved SigmaDelta Modulator Architecture

Abstract: The performance of multi-bit sigma-delta coders is analysed and compared with that which can be obtained from the much more common single-bit implementations. It is demonstrated that the fundamental difficulty with multibit designs is the requirement for a high accuracy digital to analogue converter (DAC) in the feedback loop. A general expression for the reduction of in-band noise power as a function of coder order, quantiser resolution. and oversampling ratio is derived. together with a subsidiary condition imposed by the use of multi-bit DACs in the network. A new coder topology is proposed which uses multi-bit quantisation with single-bit feedback. overcoming the manufacturing or calibration difficulties inherent in the use of high resolution DACs

Electro-mechanical position servo means

Abstract: An electro-mechanical position servo means includes a motor 10 driving a load 11 about a load axis 19, the position of the load being determined by a gyroscope arrangement 26 which provides feedback signals to close the servo loop at motor driver 12. The gyroscope 26 comprises a rate integrating gyro 13 including load angle pick-off 22 and torquer coil 21'. The pick-off output is fed back by way of amplifier 41 to torquer coil input 27 to form a gyro feedback loop, the inherent integration effected by torquer coil causing the pick-off signal, and gyroscope output signal, at 27 to be in the form of a rate or velocity signal. Two servo feedback paths are formed, a velocity feedback path 28 feeding directly to a summing junction 30 before the motor driver 12 and a position feedback path 31, 35 to the summing junction 30 including an integrator 33 of the gyroscope signals and an input summing junction 32 by which position demand signals are input at 36 as a function of position rate of change. Instead of a rate integrating gyro 13 with its gyro feedback loop the gyroscope 26 may comprise a rate gyro.