Preserving Accuracy and Enhancing Reliability in Data Acquisitions with Isolated Precision Signal Chains

Abstract

This article explores an isolated precision signal chain reference design solution and its profound impact on preserving accuracy and enhancing reliability in data acquisition applications.

Introduction

The digital age has shifted the paradigm in bringing intelligence to the edge to solve novel and complex challenges. Right at this edge sits the core of this technology—data acquisition (DAQ) systems. In the realm of data acquisition, precision and reliability are paramount. To ensure the highest level of accuracy and integrity, the implementation of an isolated precision signal chain has emerged as a critical component.

Understanding Isolated Precision Signal Chain

An isolated precision signal chain refers to a system or circuitry designed to achieve precise and accurate signal acquisition and processing while maintaining electrical isolation from the surrounding environment. Isolation is often included as part of the series of signal conditioning stages for two main purposes: safety and data integrity.1 The isolation similarly includes the following advantages:

  • Noise and interference reduction: By employing isolation techniques such as galvanic isolation, which uses transformers or optocouplers, the signal chain can eliminate common-mode voltage variations, ground loops, and electromagnetic interference (EMI). This isolation prevents external noise sources from corrupting the acquired signal, ensuring cleaner and more accurate measurements.
  • Ground loop elimination: Ground loops can introduce voltage differentials that distort the measured signal. Isolation techniques break the ground loop path, effectively removing the interference caused by varying ground potentials, thereby improving measurement accuracy.
  • Safety and protection: Isolation barriers provide electrical safety by preventing hazardous voltage spikes, transients, or surges from reaching sensitive measurement components. This protects both the measurement circuitry and the connected devices, ensuring safe and reliable operation. Aside from circuit protection, isolation also eliminates the electrical hazards for both end users and designers working on the system.

Additionally, an isolated precision signal chain consists of a series of components and techniques that work together to ensure accurate measurement and data integrity. The key components of an isolated precision signal chain typically include precision amplifiers, isolation barriers, filtering elements, and high resolution analog-to-digital converters (ADCs). These components work in tandem to eliminate noise, minimize interference, and provide accurate signal representation. An example of an isolated precision signal chain utilizing these key components is shown in Figure 1. This precision platform is a single-channel, fully isolated, low latency data acquisition system. This solution combines PGIA signal conditioning, digital, and power isolation within a compact board. The succeeding sections will discuss each of the blocks in detail, including their corresponding performance and their advantages as compared to the nonisolated equivalent.

Figure 1. Simplified block diagram of a single-channel, fully isolated, low latency data acquisition system, ADSKPMB10-EV-FMCZ.

Data and Power Isolation

The Pmod™-to-FMC interposer board houses digital isolators, regulators, and a transformer to implement galvanic isolation. Galvanic isolation is a design technique that separates electrical circuits to eliminate stray currents. Signals can pass between galvanically isolated circuits, but stray currents, such as differences in ground potential or currents induced by AC power, are blocked.2

First, data isolation is enabled by the ADuM152N and ADuM120N 3 kV rms digital isolators. These digital isolators exhibit high common-mode transient immunity (CMTI) and are highly robust to radiated and conducted noise while providing low propagation delay and low dynamic power consumption. These isolation devices, besides being easy to implement, offer exceptional performance characteristics vs. typical alternatives such as optocouplers. Specifically, the maximum propagation delay is 13 ns with a pulse width distortion of less than 5 ns. Channel-tochannel matching of propagation delay is tight at 4.0 ns and 3.0 ns maximum, respectively.

Subsequently, an isolated precision signal chain would likewise require an isolated power circuitry that satisfies the operating requirements of the signal chain. The isolated power circuitry should not affect the performance of the precision signal chain. Low emissions must be ensured from the power circuitry while also being efficient and conforming to the required safety requirements.

A good choice for power isolation is the LT3999 low noise, push-pull DC-to-DC driver, which features 1 A internal dual switches with programmable current limit, adjustable switching frequency from 50 kHz to 1 MHz (which can also be synchronized to an external clock), a wide operating input range from 2.7 V to 36 V, and a shutdown current of less than 1 μA. The push-pull topology is simple to design and implement, uses few components, and operates with low radiated emissions as an outcome of its inherently symmetrical topology.

As shown in Figure 2, the power circuity on the reference platform was designed such that the 12 V from the FMC connector can provide the power needed by the data acquisition board while enabling isolation. To achieve those, the LT3999 on the circuit drives a Pulse Electronics PH9085.083NL 2.5 kV rms isolation power transformer.

Figure 2. Reference platform isolated power circuit block diagram.

The LT3999 power converter produces an output voltage that is unregulated. The output voltage decreases with increasing load as shown in Figure 3.

Figure 3. LT3999 push-pull converter output voltage regulation.

A low dropout, linear postregulator (ADP7105) is included on the reference platform as an option to use a regulated 3.3 V output if desired. Hence, galvanically isolating the entire measurement or data acquisition circuitry through the interposer board minimizes the impact of common-mode voltage variations and external noise sources. This proves the method is an accurate, economical, and efficient way to employ an isolated measurement circuitry.3

Preserving Accuracy

Aside from implementing isolation techniques, the building blocks inside the signal chain must be matched well. Each component contributes to the entire signal chain performance and is critical in preserving the accuracy of the whole system.

Precision amplifiers—characterized by their high accuracy, low noise, and low offset voltage—provide accurate signal conditioning and amplification to ensure that the acquired signal is faithfully represented without introducing additional distortion or offsets. Furthermore, filtering elements such as low-pass filters are often employed to attenuate high frequency noise and unwanted signals, allowing only the desired signal to pass through the signal chain. This further enhances the accuracy and integrity of the measured signal. Finally, high resolution ADCs are used to convert the analog signal into a digital format for further processing or analysis. These ADCs have high sampling rates and excellent resolution, allowing for precise and detailed digitization of the analog signal. All these components are handpicked to achieve the desired performance of the reference platform.

Delving into the specifics, the data acquisition board within the reference platform showcases a discrete programmable gain instrumentation amplifier (PGIA) composed of several components, including:

  • ADA4627-1: high speed, low noise, low bias current, JFET operational amplifier
  • LT5400: precision quad matched resistor network
  • ADG1209: low capacitance, 4-channel, ±15 V/+12 V iCMOS® multiplexer
  • Internal fully differential amplifier (FDA) ADC driver of the ADAQ4003

The PGIA at the front end offers high input impedance that allows direct interface with a variety of sensors. A programmable gain is often needed to adapt the circuit to different input signal amplitudes—unipolar or bipolar and single-ended or differential with varying common-mode voltages. The PGIA works with the ADAQ4003, an 18-bit, 2 MSPS, μModule® data acquisition solution. Figure 4 illustrates the entire signal chain of this reference platform.

Figure 4. Precision medium bandwidth signal chain.

In order to verify the static performance of the reference platform, the integral nonlinearity (INL) and differential nonlinearity (DNL) were measured, respectively. Figures 5 and 6 illustrate the DNL and INL errors vs. code across various gains. The DNL errors have typical deviations of ±0.6 LSB, denoting a monotonic transfer function with no missing codes. Meanwhile, the INL errors have typical deviations of ±2.097 LSB with a visible S-shape, indicating the strong predominance of oddordered harmonics.4 These graphs depict that sufficient linearity is obtained from the entire signal chain.

Figure 5. DNL vs. code for various gains, VREF = 5 V.
Figure 6. INL vs. code for various gains, VREF = 5 V.

Employing precision amplifiers, signal conditioning techniques, and high resolution ADCs in the signal chain minimizes signal distortion, offsets, and nonlinearities, resulting in highly accurate measurements. The galvanic isolation techniques discussed earlier further reduce common-mode voltage variations and eliminate ground loop effects, ensuring an accurate representation of the measured signal.

Minimizing Noise and Interference

Noise and interference are also common challenges in data acquisition, which either emanate from components or external sources. An isolated precision signal chain addresses these issues by employing robust isolation barriers, shielding, grounding, and filtering techniques. Noise reduction techniques are incorporated in the μModule ADAQ4003 itself, enabling a high fidelity signal capture.

In particular, a single-pole, low-pass RC filter is placed between the ADC driver output and the ADC inputs inside the μModule device—which serves the following purposes: (1) eliminates high frequency noise, (2) reduces the charge kickbacks from the input of the internal SAR ADC, and (3) maximizes settling time and input signal bandwidth.5 The layout of the μModule device also ensures that the analog and digital paths are separated to avoid crossover of these signals and ease the radiating noise.

Although there are many dynamic parameters correlating to the performance of a certain data acquisition system, only three will be discussed in this article.

Dynamic range is defined as the range between the noise floor of a device and its specified maximum output level,6 which is essential in determining the smallest voltage increment that is not affected by the noise. This parameter is tested using a 5 V reference with inputs shorted to ground, at an output data rate of 2 MSPS. Figure 7 shows the dynamic range across various gains, with a typical value of 93 dB (at the highest gain setting) and 100 dB (at the lowest gain setting). Increasing the oversampling ratio to a factor of 1024× further improves the measurement, reaching up to a maximum of 123 dB and 130 dB, respectively.

Figure 7. Dynamic range.

Using the formula in Equation 1 to solve for the equivalent referred-to-output total noise would result in as low as 1.12 μV rms (OSR = 1024× at the lowest gain setting). Thus, higher dynamic range measurements signify low overall system noise.

Equation 1

Similarly, parameters such as signal-to-noise ratio (SNR) and total harmonic distortion (THD) were acquired, with a –0.5 dBFS sinusoidal signal applied at both inverting and noninverting inputs of the device. First, SNR is defined as the ratio of the rms signal amplitude to the mean value of the root-sum-square (rss) of all other spectral components, excluding harmonics and DC.7 This this can be further understood through Equation 2.

Equation 2

On the other hand, total harmonic distortion is defined as the ratio of the rms value of the fundamental signal to the mean value of the root-sum-square of its harmonics (generally, only the first five harmonics are significant),7 as depicted in Equation 3.

Equation 3

Figures 8 and 9 show the respective SNR and THD values across various gain settings. The entire signal chain achieves a maximum SNR of 98 dB and THD of –118 dB. However, these parameters degrade at high input frequencies and high gain settings. An example FFT is likewise displayed in Figure 10. The isolated signal chain displays a flat noise floor at around 140 dB below full scale and spurs buried below it. This denotes that the signal chain has good signal strength with a clean and comparable noise performance vs. its nonisolated equivalent.

Figure 8. SNR.
Figure 9. THD.
Figure 10. Single-capture FFT, fully differential input, −0.5 dBFS, 1 kHz sine.

Applications and Impact

The impact of an isolated precision signal chain spans across various industries and applications. In scientific research, it enables precise measurements in fields such as physics, chemistry, and biology, where accuracy and repeatability are paramount. In industrial automation, the signal chain ensures accurate process control, quality monitoring, and equipment diagnostics. In medical applications, it enables precise monitoring of physiological signals and accurate diagnosis. The impact also extends to areas such as environmental monitoring, energy management, and telecommunications, where reliable data acquisition is crucial for decision-making and optimization.

Floating DAQ Systems

Floating DAQ systems are best used for electronic test and measurement (ETM) applications. A common voltage measurement involves two reference points: high and low/zero potential (referred to as the earth ground). However, using the earth ground as a reference creates risky high voltage measurements. Signals with high common-mode voltages are harmful to the individual components in the signal chain. This may result in damaging both the equipment and the data. High voltages also impose hazards on the person using the equipment. Moreover, noise, coupling, and interference introduced by ground loops also raise concerns for earth grounded systems.8

Floating DAQ addresses these risks by having a separate reference floating ground point. Floating measurements enable shorter paths to the point of acquisition, and, at the same time, allow the acquisition of signals with common-mode voltages. The different ground pins in the board can be observed in Figure 11.

Figure 11. ADSKPMB10-EV-FMCZ board with different floating ground reference points.

Conclusion

An isolated precision signal chain has a profound impact on data acquisition: preserving accuracy, minimizing noise and interference, and ensuring data integrity. Incorporating precision amplification, isolation techniques, high resolution ADCs, and low noise, low emission power management enables precise measurements, even in challenging environments. The impact of an isolated precision signal chain extends to various industries, enabling advancements in scientific research, industrial automation, healthcare, and beyond. As the demand for accurate and reliable data acquisition continues to rise, the significance of an isolated precision signal chain becomes increasingly evident in driving innovation and unlocking the full potential of data-driven applications.

参考电路

1 Jen Lloyd. “Digital Isolation Technology: Putting Safety and Data Integrity First, Last, and Always.” Analog Devices, Inc.

2 Galvanic Isolation. Analog Devices, Inc.

3Digital Isolators Simplify Design and Ensure System Reliability.” Analog Devices, Inc., June 2012.

4INL/DNL Measurements for High-Speed Analog-to-Digital Converters (ADCs).” Analog Devices, Inc., November 2001.

5 Maithil Pachchigar. “μModule Data Acquisition Solution Eases Engineering Challenges for a Diverse Set of Precision Applications.” Analog Devices, Inc., November 2020.

6ADC and DAC Glossary.” Maxim Integrated, July 2002.

7 The Data Conversion Handbook. Analog Devices, Inc., 2005.

8Breaking Ground Loops with Functional Isolation to Reduce Data Transmission Errors.” Analog Devices, Inc., December 2011.

9 Van Yang, Songtao Mu, and Derrick Hartmann. “PLC DCS Analog Input Module Design Breaks Barriers in Channel-to-Channel Isolation and High Density.” Analog Dialogue, Vol. 50, No. 12, December 2016.

作者

Lloben Paculanan

Lloben Paculanan

Lloben Paculanan是ADI菲律宾GT公司的产品应用工程师。他于2000年加入ADI公司,先后担任多个测试硬件开发和应用工程职位;一直从事精密高速信号链µModule开发。他拥有美国泽维尔大学Ateneo de Cagayan学院工业工程技术学士学位,以及Enverga University的计算机工程学士学位。

Chelsea Faye Aure

Chelsea Faye Aure

Chelsea Faye Aure is a product applications engineer at Analog Devices. She provides technical applications support and works on the precision signal chain µModule® solutions development team. She started her career at ADI Philippines in 2022. She received her bachelor’s degree in electronics engineering from De La Salle University—Dasmariñas.

Jan Michael Gonzales

Jan Michael Gonzales

Jan Michael Gonzales is a product applications engineer for power systems at Analog Devices Philippines. He joined ADI in 2020 and is primarily working on powering precision signal chains. He received a bachelor’s degree in electronics engineering and a postgraduate degree in power electronics both from Mapúa University in Manila.