In part 4 of this series, we took a close look at some solutions that combine an ADC with a multiplexer. We looked at both SAR and Δ-Σ devices. We'll continue with a closer look at the Δ-Σ-plus0multiplexer version. Then we'll get an overview regarding number of bits and conversion speed with respect to various applications.
Figure 1 shows that the AD7176-2 achieves the maximum resolution up to ~22 bits, which is limited by the multiplexer switching rate and type of digital filter (Sinc3 or sinc5+Sinc1).
AD7176-2 resolution vs. switching rate.
As shown in Figure 2, Analog Devices' medium and high-speed (up to 10 MSPS) SAR ADCs with a resolution of up to 18 bits are typically used in industrial, instrumentation, and medical imaging applications that offer scalable power with throughput rate. Medium and low-speed (up to 100 ksps) Δ-Σ ADCs with a resolution of up to 24 bits are popular in industrial automation and process control for precision measurements.
Analog Devices' precision ADC portfolio.
The common issues faced by both types of ADCs in the multiplexed applications are bandwidth, settling time, and input range requirements. When you switch the multiplexer from one channel to another in a DAS, you may have large voltage amplitude steps (as opposed to pure sine waves). This can happen when one input is at a negative full-scale voltage (or sometimes ground) and the subsequent channel is at positive full-scale voltage, or vice versa.
In other words, you get a large step between the input channels in a small amount of time. You must then be concerned about the large signal bandwidth of the amplifier to settle the large voltage step. This puts an additional burden on the amplifier. With a large amplitude step, nonlinear effects appear, and slew rate and output current characteristics limit the ADC driver capabilities and output response. Some customers use the amplifier's disable feature to reduce system power or to implement a simple channel multiplexing operation.
For example, for digital X-ray imaging applications, the excellent linearity and low noise offered by the Analog Devices SAR converters enhance the image quality. A high throughput allows a shorter scanning period (more frames per second) and decreased exposure to the X-ray dosage. This leads to more accurate physician diagnostics and a better patient experience.
Multiplexing multiple channels creates higher-resolution images for full analysis of organs such as the heart, and it achieves affordable diagnosis by significantly reducing the cost while minimizing power consumption to meet the thermal constraints. Some of the high-performance SAR ADCs, such as the 18/16-bit, 5-MSPS AD7960/61, provide increased bandwidth, high accuracy, and discrete sampling in the small time window required for these types of high-performance DASs.
These converters offer the lowest noise floor relative to the full-scale input floor, resulting in a higher SNR and excellent linearity performance. The 5-MSPS fast throughput rate and low power with a small package size help designers meet space, thermal, power, and other key design challenges common to high-channel-density systems.
Accuracy, cost, power dissipation, size, complexity, and reliability are of paramount importance for medical equipment manufacturers. The medical equipment used for computed tomography (CT) and digital X-ray multiplexes multiple channels at high sampling rates into a single ADC that requires higher sampling rates without sacrificing accuracy. In CT scanners, the pixel current is captured continuously using an integrator and track-and-hold per channel, with outputs multiplexed to a high-speed ADC.
The key challenge with SAR ADCs is that they need amplifiers to drive their inputs, whereas Δ-Σ ADCs suffer from latency or settling time of filters. The SAR and Δ-Σ ADCs have their own pros and cons. The designer of the DAS must make the tradeoffs based on needed performance, speed, number of channels (space), and power. Lastly, cost requirements in a particular application may dictate the use of SAR versus Δ-Σ ADCs and whether to use external or integrated multiplexed solutions.