“This online discussion mainly introduces the principle and application of TI’s high Delta-Sigma A/D converter. The characteristic of Delta-Sigma converter is to transfer most of the noise from dynamic to resistance state, usually Delta-Sigma converter It is used in low frequency occasions where cost and requirements are required. This article will first give an overview of TI’s high Delta-Sigma A/D converter, and then introduce noise measurement and chip ADS1232.
“
This online discussion mainly introduces the principle and application of TI’s high Delta-Sigma A/D converter. The characteristic of Delta-Sigma converter is to transfer most of the noise from dynamic to resistance state, usually Delta-Sigma converter It is used in low frequency occasions where cost and requirements are required. This article will first give an overview of TI’s high Delta-Sigma A/D converter, and then introduce noise measurement and chip ADS1232.
Overview of Delta-Sigma Converters
The Delta-Sigma converter is a 1-bit converter that converts the analog Voltage into a digital quantity by means of oversampling. It consists of a 1-bit ADC, a 1-bit DAC and an integrator, as shown in Figure 1. The advantages of Delta-Sigma converters are low cost and high resolution, and they are suitable for production in today’s low-voltage semiconductor industry.
The working principle and application analysis of Delta-Sigma converter and ADS1232 chip
Delta-Sigma converter composition
The Delta-Sigma converter is composed of a differential amplifier, an integrator, a comparator and a 1-bit DAC. The input signal is subtracted from the signal from the 1-bit DAC and the result is used as the input of the integrator. When the system is in a stable working state, the integrator’s The output signal is the sum of all error voltages, and the integrator can be regarded as a low-pass filter, which has a -6dB suppression ability to noise. The output of the integrator is converted with a 1-bit ADC, and the comparator then outputs a bit stream of digital ones and zeros. The DAC converts the output of the comparator stage to a digital waveform, which is fed back to the differential amplifier.
Delta-Sigma Converter Principle Details
The integrator shapes the noise by stretching the quantization noise across the entire frequency bandwidth, while the filter filters out most of the shaping noise. There are several sources of error that reduce the effectiveness of the overall system. In order to meet the input range of the ADC, many signals require some amplification and level shifting circuits. Sometimes the amplifier is internal to the ADC, and sometimes an external amplifier is used. In either case, amplifier voltage, voltage drift, input bias current, or sampling noise will introduce an error signal. In order to obtain good ADC conversion results, the error of the amplifier should be eliminated or reduced by adjustment. The integrator has a built-in low-pass filter on the input low frequency or DC signal, which greatly reduces the noise in the channel.
The noise of a typical semiconductor amplifier is divided into two parts, 1/F noise and ground noise. The main application of Delta-Sigma ADC is in low frequency occasions, so the influence of 1/F noise dominates. 1/F noise can be controlled by choosing the right amplifier. It can be seen from the noise spectrogram (see Figure 2) that the noise of the device is mainly background noise at high frequencies, and is mainly 1/F noise at low frequencies. When it is closer to the DC signal we want to get, the greater the 1/F noise . People often think of 1/F noise as drift, which is a very low frequency phenomenon that is commonly solved by using a narrow-wave input.
The method of obtaining a narrow-band stable input is shown in Figure 3. If a 1mV radio modulation voltage is applied to the non-inverting input terminal of the differential amplifier, the 1mV signal appears at the positive output terminal, and in the following circuit, the 1mV voltage The signal is output to the negative output terminal. Since it is applied alternately to the positive and negative outputs, the result is that after averaging, the 1mV shot voltage does not appear at the output, which has a significant effect in Delta-Sigma converters. Because the output of the differential amplifier is just averaged by the integrator, the drift varies with time and pitch. For narrow-wave stabilized circuits, the actual value of pitch is irrelevant, so drift and pitch over time will not affect the conversion. the result of.
Figure 4 shows the time domain variation of a 4-bit ADC converted to a full-scale sine wave. The ADC samples the input of a sinusoidal signal, and if this signal is presented with a DAC, the effects of sampling and quantization will be easily noticed. Sampling means that the output signal is captured at a discrete point in time, and the output remains unchanged between these two points. The rate at which the input is sampled is the well-known sampling frequency. The Nyquist principle stipulates that the sampling must be at least the input Twice the signal bandwidth, and the sampling rate higher than this requirement is oversampling. Delta-Sigma uses oversampling to complete signal conversion, and the function of quantization is to convert the amplitude of continuous analog signals into discontinuous ones. level.
Oversampling reduces the level of background noise by distributing quantization noise over a wider frequency range. The Delta-Sigma converter limits the noise bandwidth by relying on a digital filter after the 1-bit ADC. Since most of the noise cannot pass through the digital filter, the effective noise of the bandwidth is reduced. The technique of distributing the quantization noise in a wider frequency range and then filtering out most of the noise with a filter is the basis for the Delta-Sigma converter to apply a low-resolution ADC.
Noise measurement
Different methods can be used to measure the noise performance of the system, and the system noise can also be expressed in different ways. It has the characteristics of a Gaussian distribution. The signal-to-noise ratio SNR is usually used for high-speed ADC systems, while ENOB is usually used for low-frequency and DC systems.
Gaussian distribution
Random noise generally has the characteristics of Gaussian distribution, and most of the sampled values will be distributed in the relevant area. If a measurement system requires a peak-to-peak limit, then 99.9% of the samples should be distributed in this area, as shown in Figure 5. Show.
peak-to-peak noise
Effective noise tells us that the sampled values are random, so it is not clear what the displayed result will be. If a displayed number of bits cannot be changed, we call it a noise-free code. Peak-to-peak noise is a statistical measure of a large amount of data, it cannot be calculated directly, it is 6.6 times the effective noise.
standard deviation
The standard definition of standard deviation requires calculating the root mean square of the difference between each measured value and the average of all measured values (as shown in Equation 1). Since the average cannot be calculated until all values are sampled, in practice In the data acquisition system of , its standard definition is not often used. An easy way is to calculate the standard deviation, which requires only two numbers, the sum of all values and the sum of squares of all numbers (as shown in Equation 2).
Calculation method of ENOB
There are two calculation methods for ENOB, the first is SNR=6.02N+1.76dB, ENOB=(SNR-1.76dB)/6.02; the second method is 2ENOB=full scale value/RMS noise value=224/. (The signal-to-noise ratio refers to the ratio of the rms value of the signal to the rms value of the noise).
ADS1232 Features and Applications
Introduction to ADS1232
The ADS1232 is a precision 24-bit AD converter with a low-noise programmable precision amplifier, a precision Delta-Sigma AD converter and a built-in oscillator. ADS1232 provides a complete front-end solution for bridge sensor applications and weighing instruments. It has very low noise. When PGA=128 times, only 17nVrms effective noise in the 20mV input range. The sampling rate is 10Hz and 80Hz, with more than 100dB rejection capability for 50Hz and 60Hz.
For weighing instrument applications, the ADS1232 is easy to use.
: It has a complete front end and does not require an external amplifier circuit.
Second it does not require an external clock.
Third, all functions are controlled by pins, and no registers need to be programmed.
In addition, the reference design of the weighing instrument can be evaluated through the EDM board of the ADS1232.
The ADS1232 provides a low drift, low noise programmable gain instrumentation amplifier consisting of 2 op amps and 3 precisely matched resistors R1, RF1 and RF2. Its selectable gains are 1x, 2x, 64x and 128x.
In weighing instruments, a large number of proportional measurement methods are used, where the ground voltage of the bridge circuit is also the reference voltage of the AD converter, because the output of the bridge circuit is proportional to the ground voltage of the bridge circuit, and the result of the AD converter is also proportional to Therefore, when the ratio method is used to measure, the output result of the AD converter is only related to the change of the bridge impedance, so the measurement can be greatly improved.
Figure 6 shows the application of the ADS1232 in a weighing instrument, where the ADS1232 has a magnification of 128 times and a data rate of 10 times/second.
Other related devices
ADS1100: 16-Bit Low-Power Converter
The ADS1100 is a 16-bit ADC converter in a SOT 23-6 package. The built-in gain can be selected between 1x, 2x, 4x or 8x, and its data rate is 8~128 times/sec. Typical applications include: Handheld devices and monitors, battery management, consumer products and industrial process controls, and more.
ADS1112: Multichannel 16-bit ADC
The ADS1112 is a 16-bit precision analog-to-digital converter with autocorrection with two differential input channels or three single-ended inputs. Built-in 2.078V voltage reference, its power supply voltage is 2.7~5.5V. Its main feature is a complete small data acquisition system, input multiplexer, PGA and oscillator. It supports an I2C interface, and typical applications include handheld devices, portable monitors, and power management.
ADS1222: 24-Bit Low-Power Converter
The ADS1222 is TI’s cost-effective 24-bit industrial Delta-Sigma converter and the industry’s leading two-channel differential input converter with high input impedance, built-in temperature sensor, two-wire serial input interface and self-calibration circuitry. Its data rate is 240SPS, typical applications include: handheld equipment and industrial process control.
ADS1271: 24-Bit High-Performance Converter
The ADS1271 is a unique high-performance 24-bit Delta-Sigma converter that combines DC and AC performance. Generally, industrial Delta-Sigma converters use high-order low-pass filters to obtain good DC, but limit the signal bandwidth and therefore only suitable for DC measurements. Whereas high-resolution ADCs for audio applications require large available bandwidth, but dc can be degraded, the ADS1271 combines excellent dc and ac performance. Typical applications include: pressure sensors, test and measurement, etc.
