A/D converter and D/A converter employing a rodix smaller than a

Details A/D converter and D/A converter employing a rodix smaller than a A/D converter and D/A converter employing a rodix smaller than a

1993-03-01

1994-07-19

Hoff, Marc S.

Coded data generation or conversion

Analog to digital conversion followed by digital to analog...

341120, 341126, H30M 100

Patent

active

053313216

DESCRIPTION:

BRIEF SUMMARY
FIELD OF THE INVENTION

The present invention relates to A/D and D/A converters and converter systems and their calibration methods.


BACKGROUND OF THE INVENTION

Digital-to-analog (D/A) and analog-to-digital (A/D) converters are used generally in electronics, especially in computer interfaces, to convert an analog signal into digital form, and vice versa. In rapid and accurate converters, it is customary to use a binary weight system in which the base number is 2: the converter, and B.sub.1 . . . B.sub.n are binary values (0,1) C is a constant and n is the number of the bits/weights of the converter.
The binary weights 2.sup.-k of the converters are usually realized by weight or reference elements forming a so-called weighing network. In practice, the weighing network and the weight elements are usually realized by means of resistors, capacitors and transistors. The accuracy of the converter thereby depends on how accurately the relationship between the values of the weight elements is the right one. This involves very high requirements for the converter manufacturing technique, and in many cases converters have to be tuned after manufacture. Further, as the values of the weight elements vary in time (drifting), the accuracy limit of converters is usually about 12 bits. Converters having a greater accuracy than that have to be calibrated or tuned intermittently during use.
Converters usually have to be calibrated by adjusting the values of the weight elements and the reference voltages.
In addition, when great speed and accuracy are to be achieved, the time required for the converter to settle to the required accuracy causes problems. For this reason, the most accurate conventional converters are usually also slow.


SUMMARY OF THE INVENTION

The object of the present invention is to provide a novel A/D or D/A converter which is more accurate, easier to calibrate and less demanding in view of the manufacturing technique than previously.
This is achieved by means of an A/D or D/A converter which according to the invention is characterized in that the base number of the number system used in the converter is smaller than 2.
The invention is based on the Applicant's discovery that when a number system having a base number smaller than 2 is used in the converter in place of the binary system, the obtained converter comprises number ranges in which the same number is formed by different combinations of bits. Consequently, deviation can be allowed for the weights without the risk of gaps remaining in the number range of the converter. The actual accurate value of the weight can be determined by calculation or programmatically with a certain self-calibration procedure and each conversion result can be corrected on the basis of these actual weights. For example, the offset and noise of the comparators of a weighing A/D converter do not cause an error in the conversion, as the same voltage value can be represented at the threshold value points of the comparators with two different bit series in the number system according to the invention. In a normal binary system, inaccuracy in the weights causes a gap in the number range of the converter, and the error caused by the gap is impossible to correct.
By means of the invention, the weights and other parameters of the converter, such as resistor and capacitor values, comparator offset, gain, rising time, etc., can be allowed to deviate fairly largely within certain limits. The structure of the converter is simple, requiring no greater accuracy from the manufacturing technique, which, in turn, reduces costs.
The speed of the A/D converter can be increased as the sampling frequency is constant and thus the amplifier need not reach its final state before conversion, because such settling time errors will be compensated for in calibration.
Converters can be calibrated (the weight values can be measured) by software at desired moments. Subsequent errors occurring in conversions are caused by noise and interferences only. The bit number of the converter can be increased up

REFERENCES:
patent: 3216001 (1965-11-01), Hinrichs
patent: 4336526 (1982-06-01), Weir
patent: 4903023 (1990-02-01), Evans et al.
patent: 4962380 (1990-10-01), Meadows
patent: 5164726 (1992-11-01), Bernstein et al.
Digital Circuits and Devices, Teuvo Kohonen, Prentice-Hall Inc., 1979, pp. 13 and 14.

Affiliated with

Also associated with

Rating

A/D converter and D/A converter employing a rodix smaller than 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 A/D converter and D/A converter employing a rodix smaller than a, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and A/D converter and D/A converter employing a rodix smaller than a will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-523076