Coded data generation or conversion – Analog to or from digital conversion – Analog to digital conversion
Reexamination Certificate
2003-03-12
2004-12-21
JeanPierre, Peguy (Department: 2819)
Coded data generation or conversion
Analog to or from digital conversion
Analog to digital conversion
C341S172000
Reexamination Certificate
active
06833803
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to analog-to-digital converters. It is disclosed in the context of analog-to-digital converters for use in power measuring instruments. However, it is believed to be useful in other applications as well.
BACKGROUND OF THE INVENTION
There are many variations on analog-to-digital (hereinafter sometimes A/D) conversion techniques Most implementations can be placed in one of six categories. These are successive approximation, flash, voltage-to-frequency, dual slope, charge balancing, and delta-sigma.
Successive approximation converters generally employ a sample-and-hold circuit, a comparator, a digital-to-analog converter and some control logic. The input signal is first captured by the sample-and-hold circuit and then a search pattern is executed using the digital-to-analog converter and the comparator. For optimization of speed, the search pattern is usually a binomial type. The input signal is scaled to be somewhere within the range of the output of the digital-to-analog converter. In the binomial search pattern the digital-to-analog converter is set to half scale and the comparator is used to determine if the captured input signal is higher or lower than the output of the digital-to-analog converter. This eliminates half of the possible results and thus determines the most significant bit of the conversion. The digital-to-analog converter is then reset to bisect the remaining voltage range and the comparator is again used to determine in which half the input voltage resides. This determines the next most significant bit. The process is repeated until the number of bits required is achieved. A twelve-bit converter requires twelve such comparisons.
Flash converters make use of a divider ladder, multiple comparators and decode logic to perform the A/D conversion. There are as many comparators and taps on the divider ladder as there are codes in the A/D converter. An 8-bit converter requires 256 comparators and 256 taps on the divider. A 12-bit converter, if produced, would require a staggering 4096. The comparators then compare the incoming signal against their respective tap voltages. Comparators with tap voltages above the input voltage assume a first state. Those with tap voltages below the input voltage assume a second state. The outputs of all the comparators are fed into the decode logic to create the output. Because they perform all the comparisons at one time, flash converters are generally considered the fastest of these six kinds of A/D converters.
Dual slope converters are a form of integrating converter. They work by measuring charge accumulated in a capacitor. If not already a current, the input signal is converted to a current and applied to a discharged capacitor for a fixed period of time. An operational amplifier, hereinafter op-amp, -based integrator circuit is frequently used to provide an extremely low burden to the input current source. Since current multiplied by time is charge and the charging time is fixed, the charge that is placed in the capacitor is proportional to the average input current. As the charge is applied, the voltage of the capacitor ramps up. This is the first slope to which the name dual slope converter refers. Next, the second step of measuring this charge is conducted. To measure the charge accumulated, the charging process is ended, and a calibrated discharging current is applied. The time required to return the capacitor to the discharged state is measured. As the charge is removed, the voltage across the capacitor ramps back down to zero. When the capacitor voltage returns to zero, exactly the amount of charge which resulted from the input current has been removed. This is the second slope to which the dual slope name refers. Since the discharge current applied and the time it was applied are both known, the charge that was removed from the capacitor, and therefore the charge that accumulated in the capacitor resulting from the input signal, is also known. If this charge is then divided by the time required for the input current to charge it, the average input current for the measurement period is calculated.
A charge balance converter is another form of integrating converter. Charge balance converters are similar to dual slope converters, in that the input signal to a charge balance converter is a current or is converted to a current, and the charge being accumulated in a capacitor is measured. They differ primarily in how charge is measured and removed. In a charge balance converter, charge continuously accumulates in the capacitor, while being simultaneously being removed in discrete quanta. Periodically, the voltage across the capacitor is measured. If enough charge has accumulated, a packet of charge is removed. This is usually accomplished by applying a calibrated current for a specific period of time. For each sample period when a packet of charge is removed, a pulse is output by the converter. If no charge is removed during the sample period, no pulse is output. The pulses, when present, appear at periodic boundaries. The frequency of the pulses is then measured to complete the conversion.
A voltage-to-frequency converter is another form of integrating converter. Voltage-to-frequency converters are similar to dual slope and charge balance converters in that the input is a current, or is converted from a voltage to a current and the charge accumulated in the capacitor is measured. They differ from dual slope- and charge balance converters in how the charge is removed. As in charge balance converters, in voltage-to-frequency converters charge is removed in discrete quanta. Unlike charge balance converters, voltage-to-frequency converters remove the charge whenever a full quantum or packet of charge has accumulated. Thus, in voltage-to-frequency converters, charge is not removed on periodic boundaries. This causes the converter to provide an output frequency which is proportional to the applied input current. The National Semiconductor LM131 family of voltage-to-frequency converters is a good example of this type of A/D converter.
The delta-sigma converter is yet another form of integrating converter. Delta-sigma converters are a highly specialized form of the charge balance converter, but are discussed separately here. The delta-sigma converter can be considered as two components, a modulator and a digital filter. The modulator contains the converter's integrating portion and charge removal portion. The modulator effectively functions as a very high speed, 1 bit digitizer with a very unique noise spectrum. This 1 bit digitizer samples at a frequency that is several orders of magnitude higher than the frequency band of interest. Because of its unique construction, the noise spectrum it produces is non-uniformly distributed and the bulk of the noise energy is outside of the frequency band of interest. Thus, by proper filtering much of this noise can be removed. This is one function served by the digital filter. The modulator is interesting in that it can perform the voltage-to-current conversion as an inherent part of its function, and thus, from the user's perspective, the input to the converter is usually a voltage instead of a current. The digital filter performs two functions. It functions as a very sophisticated version of the counter in the charge balance converter, as well as a digital filter to extract a higher resolution result at a lower data rate than the 1 bit digitizer.
Traditional power measurement has principally revolved around measuring power flow in power delivery circuits. The measurements, whether they are watts, watt-hours, VoltAmpereSReactive (VARS), Q-hours, or the like, have usually been measured one at a time. Performing these measurements involves the precision multiplication of a voltage signal and a current signal. Traditionally this has been performed with analog circuitry. The most commercially successful of these circuits has been the pulse width modulator. In communications circuitry, it is often referred to as a balanced mixer or ring demodul
Jean-Pierre Peguy
Radian Research, Inc.
LandOfFree
Methods and apparatus for analog-to-digital conversion does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Methods and apparatus for analog-to-digital conversion, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Methods and apparatus for analog-to-digital conversion will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3313361