Digital-to-analog converters (DACs) are widely used in various applications and are often paired with amplifiers to shape the output signal. The amplifier can enhance current drive capability, convert differential signals into single-ended outputs, isolate downstream components, or provide complementary bipolar voltage outputs. Figure 1 illustrates a typical signal chain for a single-supply system, which includes a reference voltage source, a DAC, and a buffer. To ensure high dynamic range and excellent signal-to-noise ratio (SNR), DACs are typically designed to operate at full swing, with the reference voltage (VREF) set equal to the supply voltage (VDD). This setup maximizes the utilization of digital codes. When using a single power supply, both the DAC and the output buffer are usually connected to the same power rail. In this configuration, rail-to-rail input and output amplifiers are essential.
Figure 1: Typical signal chain for single-supply operation
The input stage of a traditional non-rail-to-rail amplifier usually employs a p-type or n-type differential pair. A p-type input amplifier allows the input common-mode voltage to approach the lower supply rail, making it suitable for ground detection applications. On the other hand, n-type input amplifiers can handle input voltages from a few volts above the low supply rail up to the high supply rail, making them ideal for high-side current sensing. Rail-to-rail input amplifiers integrate both n-type and p-type differential pairs to extend the common-mode input range across both supply rails.
These amplifiers use two parallel differential pairs, one n-type and one complementary p-type, to determine which pair is active based on the input common-mode voltage. The p-type pair activates when the input approaches the low supply rail, while the n-type pair becomes active for higher input voltages. This design enables the amplifier to support a wide input range and swing close to both supply rails. However, a drawback of this configuration is "crossover distortion," where the offset voltages between the two differential pairs differ. When the input common-mode voltage crosses the transition point, the amplifier may exhibit a step-like response, which can be observed in the offset voltage versus input common-mode voltage curve. This distortion is inherent in all rail-to-rail amplifiers with dual differential pair topologies.
Figure 2: The relationship between the input offset voltage of a typical rail-to-rail input amplifier and the input common-mode voltage
Figure 3 shows the integral nonlinearity (INL) error of a circuit that uses a 16-bit DAC along with a standard rail-to-rail input/output buffer. INL represents the deviation of the actual transfer function from the ideal one, measured in least significant bits (LSB). In this example, the DAC scans from code 200 to code 2¹â¶â€“200, excluding approximately 15 mV (or 200 codes) due to the limited output swing of the rail-to-rail amplifier. Crossover distortion is detected around 45,000 digital codes, corresponding to an input common-mode voltage of about 3.4 V. This distortion degrades the INL performance and reduces overall system accuracy. For a 16-bit system, the crossover-induced nonlinearity can reach as high as 4 to 5 LSB. While some systems perform calibration to eliminate initial offset, this type of nonlinearity cannot be corrected through calibration alone.
Figure 3: Integral Nonlinearity (INL) for a 16-Bit DAC with a Typical Rail-to-Rail Input Buffer
To address the crossover distortion issue, zero-crossing distortion amplifiers can be used. These amplifiers incorporate an on-chip charge pump input enhancement circuit to achieve rail-to-rail input swing without requiring complementary differential pairs. The charge pump boosts the internal supply voltage, providing sufficient headroom for the input stage and eliminating crossover distortion. The ADA4500-2 from Analog Devices is an example of such a zero-crossing distortion amplifier. Figure 4 shows the offset voltage of the device versus the input common-mode voltage. It demonstrates that the offset remains stable throughout the entire input range.
Figure 4: The relationship between the offset voltage of a zero-crossing distortion amplifier and the input common-mode voltage
Using a zero-crossing distortion amplifier eliminates crossover nonlinearities in DAC systems. Figure 5 shows the INL performance of the same 16-bit DAC with the ADA4500-2 buffer. The improved INL is less than ±1 LSB, significantly enhancing system accuracy.
Another alternative is to reduce the reference voltage (VREF) below the supply voltage (VDD), which can avoid crossover distortion. For instance, using a 5 V supply with a 2.5 V reference ensures that the crossover region of a typical rail-to-rail amplifier lies outside the digital code range. However, this method halves the output range, potentially requiring an external amplifier to boost the signal. If multiple power supplies are available, increasing the supply voltage for the amplifier can allow the use of non-rail-to-rail input amplifiers, offering more headroom for the input stage. But this approach comes at the cost of reduced power efficiency.
In summary, selecting the right amplifier as a DAC output buffer is crucial. You can either reduce the DAC reference voltage to expand the output range or increase the buffer supply voltage to improve performance, though both options have trade-offs. Alternatively, using rail-to-rail input/output amplifiers maximizes the input and output range, but it's best to choose a zero-crossing distortion amplifier to avoid crossover nonlinearities and ensure optimal system performance.
Electric Motor For Circular Knitting Machine
Electric Motor For Circular Knitting Machine,Drive Circular Knitting Machine Device,Special Motor For Loom,Dc Permanent Magnet Synchronous Motor
Jiangsu Hengchi Motor Technology Co., Ltd , https://www.hcemotor.com