JPH01319330A - AD converter - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to an oversampling type AD converter, and more particularly to an AD converter that achieves high speed and high accuracy through a parallel configuration.

Conventional technology A parallel type is superior as a high-speed A/D converter. In this method, for an n-bit AD converter, 2'-1 comparators are prepared, the voltage to be compared is applied to the comparators, and this reference voltage is compared with the input voltage to determine the level of the input voltage. It is something to detect. Although this method is fast, it requires a large number of comparators, which increases the number of elements, and it is difficult to improve conversion accuracy.

On the other hand, there is an oversampling type A/D converter as a method that does not require high accuracy in the elements used.
(For example, Nikkei Electronics No. 447 (19
88,5,16) I) IE35) The sampling frequency of a general A/D converter is set to about 3 degrees of the signal frequency band. On the other hand, oversampling type A/D converters can perform high-precision conversion even if the accuracy of the elements used is low by setting the sampling frequency to a high frequency that is several times higher than the signal frequency band. be.

FIG. 2 shows a block diagram of a conventional oversampling A/D converter. 11 is an adder, 12 is a comparator, 13 is a 1-sample delay circuit, 14 is a 1-bit D/A converter, 1
5 is an integrator, 16 is an analog signal input terminal, and 17 is an output terminal to a digital filter.

The analog input signal 16 and the predicted voltage outputted by one integrator 15 are compared, and the magnitude of the comparison is compared by 12 comparators. This comparator outputs "1" if the input voltage is larger. This signal is delayed by one sample period in the delay circuit 13,
1 pin) is input to the D/A converter 14. The output voltage of the integrator 15 increases slightly with the output of this D/A converter.

Conversely, if the input voltage is lower, the comparator 12 outputs "0", and as a result, the output voltage of the integrator 12 decreases slightly. That is, this A/D converter is fed back so that the difference between the predicted signal, which is the output of the integrator 15, and the analog input signal is less than or equal to the magnitude of the output of the D/A converter 14.
A method for processing the digital signal output at this time
/D converter.

Problems to be Solved by the Invention This oversampling type AD converter can realize a high-precision A/D converter even if the precision of the component elements is low, and it is similar to a normal successive approximation type AD converter. It does not require a sample-and-hold circuit, which is difficult to speed up. However, since it requires a high-speed sampling frequency 100 times higher than the analog input signal band, it is optimal as an A/D converter for low signal band signals such as audio, but it is suitable for A/D converters for video frequency band signals. As a device, the clock frequency becomes extremely high, making it difficult to implement.

Means for Solving the Problems The present invention provides an oversampling type A/D in which an adder, a plurality of comparators, a register, a D/A converter with precision of 2 bits or more, and an integrator are connected in a loop. In the converter, the D/A converter receives the outputs of the plurality of comparators as input, and the output of an integrator connected to the output of the D/A converter is proportional to the output of the D/A converter. This is an A/D converter characterized in that it changes as follows.

Function: According to this A/D converter, even if the precision of the elements used is low and the sampling frequency is low, granular noise and gradient overload noise can be kept low, and high speed,
Highly accurate A/D conversion can be achieved.

Embodiment FIG. 1 shows an embodiment of an oversampling type A/D converter of the present invention. 1 is an adder, 2-1 to 2-3 are comparators, 3 is a register, 4 is a 2-bit D/A converter, 5 is an integrator, 6 is an analog signal input terminal, and 7 is an output terminal to the digital filter. It is.

The analog input signal 6 and the predicted voltage output from the integrator 5 are input to the adder 1, and the difference is input to the comparator 2-1.2-2.2.
Output to -3. Comparator 2-2 outputs "1" if the input voltage is higher. The comparator 2-1 outputs "1" when the analog input signal is larger and the difference therebetween is larger than a certain value. The output of comparator 2-3 is “0”
It is.

These digital signals are delayed by a register 3 for one sampling period and input to a 2-bit D/A converter 4. The output voltage of the integrator 15 increases slightly with the output of this D/A converter. The rate of this increase depends on the magnitude of the output of the D/A converter 4.

Conversely, if the input voltage is lower, the comparator 2-2 outputs "0". The comparator 2-3 outputs "1" when the analog input signal is smaller and the difference is even smaller by a certain value. These digital signals are delayed in register 3 for one sample period. This digital signal is input to the D/A converter 4, and the output voltage of the integrator 2 is slightly decreased by the output of this D/A converter. When the difference output of adder 1 is smaller than -, the decrease in the output voltage of integrator 2 is large.

In this way, the output of the integrator 5 changes following the analog input signal, and the digital output signal of the comparator 2-2 is changed accordingly. This digital output is output from the output terminal 7 to the digital filter. A digital filter cuts components higher than the analog input signal frequency band. At this time, granular noise is removed and a high bit digital signal is replaced.

In this example, the outputs of the comparators 2-1, 2-3 are not used as inputs to the digital filter, but if this data is used, the conversion accuracy can be further improved.

In this way, the oversampling type A/D converter of the present invention realizes an A/D converter with low granular noise and low slope overload noise even at a sampling frequency that is not much higher than the analog input signal frequency band. I can do it.

Of course, in this embodiment, three comparators and a 2-bit D/A converter are used, but the present invention is not limited to this, and more comparators and a higher bit D/A converter are used. The sampling frequency can be further reduced by using

Effects of the Invention Since the oversampling type A/D converter of the present invention can lower the sampling frequency while maintaining high accuracy, even video band A/D converters can be easily integrated into semiconductors.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an embodiment of an oversampling type A/D converter of the present invention, and FIG. 2 is a block diagram of an example of a conventional oversampling type A/D converter. 1. Adder, 2. Comparator, 3. Register, 4.
D/A converter, 5...integrator.

Claims (1)

[Claims] The oversampling type A/D converter is composed of at least a plurality of comparators and a D/A converter with a precision of 2 bits or more, and the D/A converter converts the outputs of the plurality of comparators. An AD converter characterized in that the output of an integrator connected to the output of the D/A converter as an input changes in proportion to the output of the D/A converter.