Coded data generation or conversion – Converter compensation
Reexamination Certificate
1999-08-16
2001-09-04
Wamsley, Patrick (Department: 2819)
Coded data generation or conversion
Converter compensation
C341S120000, C324S318000
Reexamination Certificate
active
06285304
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention is directed to an analog-to-digital converter circuit and is also directed to a control device for a gradient amplifier of a magnetic resonance tomography apparatus. The analog-to-digital converter circuit can be utilized for all applications wherein high precision, resolution and constancy over time are required. In particular, the analog-to-digital converter circuit is provided for the digitization of an actual current value signal in a control circuit of a gradient amplifier.
2. Description of the Prior Art
In a nuclear magnetic resonance tomography apparatus, magnetic field gradients are produced by gradient coils, each of which is connected to a gradient amplifier. During the measuring event, each gradient coil has a current flowing therein that, for example, can assume values up to 300 A with a predetermined current curve shape. The shape of the current curve must be adhered to exactly up to a few mA. A complicated control circuit is required in order to achieve this precision.
In known gradient amplifiers, the control circuit including the drive electronics for a switched output stage (modulator) is implemented as an analog circuit. This, however, limits the functionality. More complex applications, for example, energy balancing between a number of output stages or a more detailed answerback of individual output stage parameters, cannot be economically accomplished in analog technology.
There is therefore a desire to design the control circuit and modulator as fully digitally as possible, for example with a suitably programmed digital signal processor (DSP). However, the problem of the digitization of the analog actual current value signal determined by a current converter with the required precision, sampling rate and stability thereby arises. A corresponding problem occurs if, instead of the actual current value signal, some other analog signal that is to be digitalized with high precision for input into a digital processing means.
A digital-to-analog converter referred to as “Tracking ADC” is described in the book “Halbleiter-Schaltungstechnik” by U. Tietze and Ch. Schenk, Springer Verlag, 10
th
Edition 1993, pages 784-785, (FIG. 23.43 and associated description). This known converter circuit has a differential amplifier which compares an analog input signal to an analog converter signal and generates an analog difference signal corresponding to the difference therebetween, a digital-to-analog converter which generates the aforementioned analog converter signal from a digital converter value, and a control unit which generates the aforementioned digital converter value as the output of the overall circuit. However, the conversion speed of this circuit as well as the zero point stability and noise suppression are low. Employment thereof for gradient amplifiers is not suggested.
SUMMARY OF THE INVENTION
An object of the present invention is to provide an analog-to-digital converter circuit, and a control circuit therefor, which avoid the aforementioned problems of the prior art and which, with a low outlay, exhibit high precision, resolution and stability.
The above object is achieved in accordance with the principles of the present invention in an analog-to-digital converter circuit having a differential amplifier which compares an analog input signal to an analog converter signal and which generates an analog difference signal according to the difference between these two signals, a digital-to-analog converter which generates the aforementioned analog converter signal from a digital converter value, a control unit which generates the aforementioned digital converter value, which is also the output value of the overall analog-to-digital circuit, an integrator which integrates the analog difference signal and which generates an analog integrator signal dependent in the integration result, and an analog-to-digital converter which generates a digital integrator value from the analog integrator signal, the control unit generating the digital converter value dependent at least on the digital integrator value.
The invention proceeds from the basic idea of determining a deviation of the analog input signal (actual value signal) from an analog converter signal (reference value signal). The analog converter signal (reference value signal) is generated by a high-precision digital-to-analog converter. The objective is to optimally minimize the identified deviation by suitable re-adjustment of the digital-to-analog converter or of the entire control circuit. Inventively, an integrator and a first analog-to-digital converter are provided in order to integrate and digitize the deviation (difference) identified by a differential amplifier before further interpretation.
The integrator acquires the difference signal emitted as an output by the differential amplifier without interruptions. A noise emission into the system is compensated by the integrator. As a result, the integrator achieves an enhanced zero point stability and a high noise suppression.
The first digital-to-analog converter digitizes the integrator output signal. By contrast to a mere comparator, a digital-to-analog converter exhibits an output word width of at least a few bits. A number of deviation classes thus can be distinguished. This enables a fast re-adjustment of the system since the step width of the re-adjustment can be selected to be longer as the absolute amount of the digitalized, integrated deviation (digital integrator value) increases.
The inventive apparatus can, for example, be designed for a resolution and precision of 18 bits given a 100 kHz sampling rate and a stability that is better than the least significant bit. Such values could be hitherto achieved, if at all, only with extremely high outlay.
The employment of an integrator offers the surprising advantage that a sample and hold circuit can be omitted. In conventionally constructed analog-to-digital converters, such a sample and hold circuit must precede the actual converter in order to avoid a falsification of the result in the case of an input voltage that changes during the digitalization process. It is thereby disadvantageous that changes of the input voltage during the hold phase are not acquired, or a brief-duration disturbance during the sampling phase is “frozen” for the entire hold phase. These two effects, that can falsify the measured result, are avoided by the elimination of a sample and hold circuit which is enabled by the invention.
In preferred embodiments, the resolution of the first analog-to-digital converter amounts to between 4 bits and 16 bits and, especially preferred, to between 8 bits and 14 bits. Given these resolution values, a adequately fine graduation of the deviation classes can be obtained with low cost outlay. Given an unchanging input signal, a level change of the digital-to-analog converter by one unit (1 LSB) preferably effects a change of the digital integrator signal by at least one or two or four units. This can be achieved by a suitable design of the converter and/or by an amplification (scaling) of the analog difference value.
The digital-to-analog converter preferably has a very high precision, since its precision directly influences the measured result. The resolution and precision of the digital-to-analog converter preferably amounts to between 16 bits and 24 bits and is at least 2 bits or at least 4 bits or at least 6 bits higher than the resolution of the first analog-to-digital converter.
In order to obtain an especially high system speed, a second analog-to-digital converter is preferably provided that emits the digitalized difference signal as an output directly to the control stage or directly to the regulator. This second analog-to-digital converter in preferred embodiments exhibits characteristics as cited above in conjunction with the first analog-to-digital converter.
Preferably, the control stage or regulator minimizes the amount of the digital integrator value and/or digital difference value. The control stage in
Schiff & Hardin & Waite
Siemens Aktiengesellschaft
Wamsley Patrick
LandOfFree
Analog-to-digital converter circuit and control device for a... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Analog-to-digital converter circuit and control device for a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Analog-to-digital converter circuit and control device for a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2495146