JPS634967B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an automatic correction method for an analog-to-digital converter, and more particularly to a correction amount measurement method that can be easily automated using a single sheet of paper.

In an analog-to-digital converter, the amount of deviation of the digital output value from the expected value is called error. These errors include quantization errors, linearity errors, offset errors, etc. In any case, the accuracy of the analog-to-digital converter is determined by the magnitude of the error, so it is necessary to correct the error.

Conventionally, analog-to-digital converters (hereinafter referred to as A/
In order to obtain the correction amount for the local digital-to-analog converter (hereinafter referred to as the local D converter), which is normally connected only to the comparator,
This was determined by outputting the output of a local DAC (called a DAC) to an external device and measuring the output voltage for this local DAC's input code with a high-precision voltmeter. However, this has the following drawbacks.

(1) A high-precision voltmeter is required, and currently this type of equipment is expensive and takes a long time to measure.
It takes a lot of time to find the chip correction amount, which reduces productivity.

(2) Errors caused by the comparator cannot be corrected.

(3) It is necessary to output the local DAC output externally.

In case (3), if the output resistance of the local DAC is low, the local DAC output can be output directly to the outside, but the wiring for external output is long and there are pads, pins, etc. Measurement accuracy deteriorates due to the noise generated. local DAC
If the output resistance of the DAC is high, or to reduce the noise mentioned above, it is possible to connect a buffer amplifier to the local DAC output and output the buffer amplifier output externally, but in this case, the error of the buffer amplifier will be included in the measured value. This will degrade measurement accuracy.

In order to eliminate these drawbacks, the present invention makes it possible to measure the correction amount by using a comparator and a sample hold circuit included in the A/D converter instead of a measuring device such as a voltmeter. The drawings are designed to simplify measurement and improve productivity.The drawings will be explained in detail below.

FIG. 1 is a diagram explaining the principle of the present invention. Here,
The digital-to-analog converter that generates the output of the upper digit will be called the upper DAC, and the digital-to-analog converter that generates the output of the lower digit will be called the lower DAC. The combination of this upper DAC and lower DAC is the local DAC. In FIG. 1, A is the characteristic of the entire local DAC, L is the characteristic of the lower DAC, M is the characteristic of the upper DAC, is the full scale of the lower DAC, and p is 1 LSB of the lower DAC. Set the digital input code of the upper DAC appropriately,
Upper and lower values when the lower DAC is set to full scale (setting digital code is D _a )
The total output voltage of the combined DAC outputs is _Vs.

(1) The voltage V _s is sampled and held in a sample-and-hold circuit, and a local
When converting from analog to digital with DAC, V _s
The other code D _b corresponding to can be found. Therefore, D _c as follows: D _b −D _a =D _c (1) is obtained for all codes of the upper DAC. This is designated as D _c0 to D _cK . Next, for any analog input voltage V _d , the local
Analog-to-digital conversion is performed using the characteristics of the DAC.
The conversion result at this time is D _f . At this time, the code of the upper DAC is m _f (upper M bits of D _f )
If D _c = _nf 〓 ^p=0 D _cp ...(2) Find D _c from the formula and subtract D _c from D _f to perform digital-to-analog conversion with ideal characteristic I. You can obtain a digital code D _g similar to .

(2) Instead of sampling and holding V _s , sample and hold V _s +1, which is 1 LSB larger than V _s , and find the corresponding digital code D _e , D _e - (D _a +1) = D _c ... A similar correction amount can be obtained by calculating the correction amount from the equation (3).

Measuring the amount of correction by this method does not require a measuring device such as a voltmeter, and has the advantage that the amount of correction can be automatically measured with a simple control and addition circuit assembled with a digital circuit.

Figures 1 and 2 show examples of the local DAC of the present invention; 1 is for sampling and holding _Vs ;
(2) is for sampling and holding V _s +1.
Circuit 1 is an example in which general capacitor array DACs are used as upper and lower DACs. (2)
The circuit should always be grounded with a capacitance of 1C V _ref
The feature is that it can be switched between the control terminal _Cb and the ground.

In both circuits (1) and (2), switch S _a is connected to ground, and S _p
~ Connecting S _e-1 to the V _fef side outputs the full scale of the lower DAC, and the circuit (2) further connects the switch S _b
Connecting to the V _ref side provides an output that is 1 LSB larger than the full scale of the lower DAC. At this time, when the digital inputs from B _{( )} to B _{( +n-1)} are appropriately set and the switch S _s is closed, the voltage of V _s or V _{s +1} shown in Figure 1 is sampled, It is held when the switch S _s is opened. In this way, analog to digital conversion is performed using a method such as successive approximation using a comparator so that the output OUT becomes the ground level, and the digital code D _b is obtained.

As explained above, according to the correction method of the present invention, it is possible to use a conventional local DAC as is, and the correction amount can be calculated using a simple digital circuit without using a measuring device such as a high-precision voltmeter. Since the measurement can be performed automatically, there is no need to increase the circuit scale, which has the advantage of improving productivity and reducing costs.

[Brief explanation of the drawing]

FIG. 1 is a diagram explaining the principle of the present invention, and FIG. 2 is a diagram showing an embodiment of a local digital-to-analog converter according to the present invention. S _p ~ _{S +n-1} ...Ground, V _ref selector switch,
S _a ...Ground, A input selector switch, S _s ...Output discharge switch, 1C~2 ^m-1 C...
Capacity, B _p ~ _{B +n-1} ...Digital input, S _b ...
Ground, V _ref changeover switch, C _a , C _b , D _s ... switch control signal terminals.

Claims (1)

[Claims] 1 The first digital signal that generates the upper digit output
an analog converter, and the first output as the output of the lower digit.
Output value of 1 bit of the least significant digit digital input of the digital-to-analog converter (1LSB value)
the second, which produces a full-scale output that is always greater than
a digital-to-analog converter, the local digital-to-analog converter is configured to add the output of the first digital-to-analog converter and the output of the second digital-to-analog converter to obtain an analog output. In the automatic correction method for an analog-to-digital converter having a converter, a digital input code is set in the first digital-to-analog converter, and the output of the second digital-to-analog converter is set to a full-scale output or a full-scale output. Increase the output voltage by 1 LSB, sample and hold the output voltage at this time, find the digital input code that outputs a voltage equal to this voltage through analog-to-digital conversion, find the difference between this code and the preset code, and do the same. Find this difference for all digital input codes of the first digital-to-analog converter, and calculate the correction amount for any input code of the first digital-to-analog converter as the sum of the correction amounts for input codes smaller than that input code. An automatic correction method for an analog-to-digital converter, characterized in that: