In this article we will learn about Data Acquisition - DAQ, describing it with enough detail that you will:
Are you ready to get started? Let’s go!
Data acquisition (commonly abbreviated as DAQ or DAS) is the process of sampling signals that measure real-world physical phenomena and converting them into a digital form that can be manipulated by a computer and software.
Data Acquisition is generally accepted to be distinct from earlier forms of recording to tape recorders or paper charts. Unlike those methods, the signals are converted from the analog domain to the digital domain and then recorded to a digital medium such as ROM, flash media, or hard disk drives.
Modern digital data acquisition systems consist of four essential components that form the entire measurement chain of physics phenomena:
Data acquisition system block diagram and the elements of the modern digital data acquisition system
The typical data acquisition system has multiple channels of signal conditioning circuitry which provide the interface between external sensors and the A/D conversion subsystem.
Data acquisition systems are principally in the business of measuring physical phenomena such as:
Note that there are several other measurands, including light and images, sound, mass, position, speed, etc. that can be measured by the data acquisition system.
Dewesoft provides easy-to-use modern and modular digital data acquisition systems for the simple and for most demanding test and measurement applications while also offering an industry-leading 7-year warranty.
The primary purpose of a data acquisition system is to acquire and store the data. But they are also intended to provide real-time and post-recording visualization and analysis of the data. Furthermore, most data acquisition systems have some analytical and report generation capability built in.
A recent innovation is the combination of data acquisition and control, where a high-quality DAQ system is connected tightly and synchronized with a real-time control system. You can read more about this topic in the related article: “Merging Data Acquisition with a Real-Time Control System”.
Engineers in different applications have various requirements, of course, but these key capabilities are present in varying proportions:
A high-quality Dewesoft data acquisition system offer a complete solution for data recording, storing, visualization, analysis, and reporting in one complete package
The data acquisition instrumentation is also heavily used in monitoring applications. Such examples are:
Data acquisition systems or DAQ devices are essential in the testing of products, from automobiles to medical devices - basically, any electromechanical device that people use.
Before data acquisition, products were tested in an unstructured, highly subjective manner. For example, when testing a new suspension in an automobile, engineers often relied on the opinions of test drivers as to how the suspension “felt” to them.
With the invention and development of data acquisition systems, which could collect data from a wide variety of sensors, these kinds of subjective opinions were replaced with objective measurements. These could easily be repeated, compared, analyzed mathematically, and visualized in many ways.
Example of a testing scenario where Dewesoft's data acquisition system is used to record, store and analyze the data during extreme weight load testing on a truck
Today, no one would consider making any kind of vehicle, large or small, aircraft, medical devices, large-scale machinery, etc without employing data acquisition to objectively measure their performance, safety, and reliability.
Check out digital and modular data acquisition systems with an industry-leading 7-year warranty:
Data acquisition is the process of converting real-world signals to the digital domain for display, storage, and analysis. Because physical phenomena exist in the analog domain, i.e., the physical world that we live in, they must be first measured there and then converted to the digital domain.
This process is done using a variety of sensors and signal conditioning circuitry. The outputs are sampled by analog-to-digital converters (ADCs) and then written in a time-based stream to a digital memory media, as mentioned above. We usually call such systems the measurement systems.
A complete scheme of an analog data acquisition system
Let’s look at each of these elements in the chain in more detail:
The measurement of a physical phenomenon, such as the temperature, the level of a sound source, or the vibration occurring from constant motion, begins with a sensor. A sensor is also called a transducer. A sensor converts a physical phenomenon into a measurable electrical signal.
Sensors are used in our everyday lives. For example, the common mercury thermometer is a very old type of sensor used for measuring temperature. Using colored mercury in a closed tube, it relies on the fact that this chemical has a consistent and linear reaction to changes in temperature. By marking the tube with temperature values, we can look at the thermometer and see what the temperature is with limited precision.
The classical thermometer is used to measure temperature for centuries
Of course, there is no analog output other than the visual one. This kind of primitive thermometer, while useful in the oven, or outside the kitchen window, is not particularly useful for data acquisition applications.
So other types of sensors have been invented to measure temperatures, such as thermocouples, thermistors, RTDs (Resistance Temperature Detectors), and even infra-red temperature detectors. Millions of these sensors are at work every day in all manner of applications, from the engine temperature shown on our automobile dashboard, to the temperatures measured in pharmaceutical manufacturing. Virtually every industry utilizes temperature measurement in some way.
Temperature sensors: from left to right: thermocouple, thermistors, RTD sensor
Of course, there are many other types of sensors that have been invented to measure another physical phenomenon:
Depending on the type of sensor, its electrical output can be a voltage, current, resistance, or another electrical attribute that varies over time. The output of these analog sensors is typically connected to the input of a signal conditioner, which we will discuss in the next section.
Signal conditioners are in the business of taking the output from analog sensors and preparing them to be sampled digitally.
If we continue the example of the thermocouple. The signal conditioning circuitry needs to linearize the output from the sensor as well as provide isolation, and amplification to bring the very small voltage up to a nominal level for digitizing.
From analog signal sources to digitalized data ready for processing by computer and software
Each signal conditioning circuitry is designed by the manufacturer to perform the elemental normalizing of the sensor output to ensure its linearity and fidelity to the source phenomena, and prepare it for digitizing. And since every sensor type is different, the signal conditioners must conform perfectly to them.
Sometimes also referred to as galvanic isolation, electrical isolation is the separation of a circuit from other sources of electrical potentials. This is especially important with measuring systems because most signals exist at relatively low levels, and external electrical potentials can influence the signal quality greatly, resulting in wrong readings. Interfering potentials can be both AC and DC in nature.
For example, when a sensor is placed directly on an article under test, (e.g. a power supply) which has potential above ground (i.e., not at 0V), this can impose a DC offset on the signal of hundreds of volts. Electrical interference or noise can also take the form of AC signals created by other electrical components in the signal path or in the environment around the test. For example, fluorescent lights in the room can radiate 400Hz which can be picked up by very sensitive sensors.
This is why the best data acquisition systems have isolated inputs - to preserve the integrity of the signal chain and ensure that what the sensor outputs is truly what has been read. There are several kinds of isolation techniques employed today.
Video explaining high galvanic isolation on Dewesoft data acquisition systems
Virtually every signal that we want to measure can be affected by electrical interference or noise. This has a variety of causes, including ambient electromagnetic fields which can be induced into high gain signal lines, or simple voltage potentials that exist between the sensor or measuring system and the object under test. Therefore, the best signal conditioning systems provide selectable filtering that the engineer can use in order to remove these interferences and make better measurements.
In this scheme, a noise analog signal is passed through a low pass filter to filter unwanted frequencies
Filters are normally expressed in terms of the band that they operate upon. There are four basic types of signal filters:
Basic filter types
Some filtering, such as anti-aliasing filtering, can only be done in the analog domain. This is because once a false signal caused by under-sampling has been digitized, there is no way to know what the real signal looked like anymore. However, nearly all other filtering can be done in the digital domain, i.e., in software, after the signal has been digitized.
Filters are also defined by how many poles they have. The more poles, the steeper the roll-off they are capable of performing on the signal. This roll-off or slope simply means how many decibels of the signal can be rolled off per octave. The specification of the filter in question will typically give the maximum roll-off in dB/Q.
Dewesoft DAQ hardware typically provides low-pass filtering as required by the types of signals being measured. Some conditioners additionally provide high-pass filtering, for example, CHARGE signal amplifiers. Removing unwanted low-frequency elements is especially critical if the measured signal will be integrated or double-integrated, as unwanted elements would badly distort the derived speed or displacement values.
You will also hear of filter types such as Bessel, Butterworth, Elliptic, and Chebyshev to name just a few. Because all filters impose distortions onto the signal itself due to their very nature, engineers over the years have developed their own filtering types in order to provide the best possible results for their specific purposes.
| Filter Type | Roll-off Steepness | Ripple or Distortion | Other Factors |
|---|---|---|---|
| Butterworth | Good | No ripple, but square waves cause distortion (hysteresis) | Moderate phase distortion |
| Chebyshev | Steeper | Ripples in the pass-band | Poor transient response |
| Bessel | Good | No ringing or overshoot from non-sinus waveforms | Increased phase delay |
| Elliptic | Steepest | Ripples in the pass-band | Non-linear phase response |
You can see that there are trade-offs among these filter types. Therefore it is up to the engineer to choose the best filter type for their application.
DewesoftX DAQ software provides a broad palette of user-selectable filtering options, including all of the ones mentioned above and more. It is interesting to note that software filters can be applied after measurement - and even removed or modified after measurement. This provides the engineer with many tools by which to analyze their data non-destructively.
Filtering setup inside DewesoftX data acquisition software
Using DewesoftX software, engineers can record their data without filtering, and then apply various filters after recording and experimenting, even doing side-by-side comparisons with the original signal. This flexibility is a powerful analysis tool and extremely easy to implement. It preserves the raw, unfiltered data and simultaneously allows the engineer to apply filters as needed, creating a different data set for analytical or presentation purposes.
The output of most physical measurement signal conditions is an analog signal. It is necessary to convert this signal to a series of high-speed digital values so that it can be displayed and stored by the data acquisition system. As such, an A/D card or A/D subsystem is used to convert this signal.
AD converter scheme - converts the analog signal into digital domain data
There are a variety of ADC types, including both multiplexed and single converters per channel. In a multiplexed ADC system, a single analog-to-digital converter is used to convert multiple signals from analog to digital domain. This is done by multiplexing the analog signals one at a time into the ADC.
This is a lower-cost approach compared to having an ADC chip per channel. But on the other hand, it is not possible to precisely align the signals on the time axis, because only one signal can ever be converted at a time. Therefore, there is always a time skew between channels.
In the early days of data acquisition, 8-bit ADCs were common. As of this writing, 24-bit ADCs are standard among most data acquisition systems designed to make dynamic measurements, and 16-bit ADCs are commonly considered the bare minimum resolution for signals in general.
The rate at which the signals are converted is called the sample rate. Certain applications, such as most temperature measurements, do not require a high rate since the measurands do not change very rapidly. However, AC voltages and currents, shock and vibration, and many other measurands require sample rates in the tens or hundreds of thousands of samples per second or more. The sample rate is considered the T or X-axis of measurement.
ADC Sampling rate
On the Y or vertical axis, ADCs are available with various resolutions. The most common today are 16-bit and 24-bit. An ADC with a 16-bit resolution can theoretically digitize an incoming signal with a resolution of one part in 65,535 (2^16 = 65,536).
This number is actually reduced by noise and quantization error among other factors, but it provides a good starting point for comparison. Because each bit of resolution effectively doubles the quantization resolution, systems with 24-bit ADCs provide 2^24 = 16,777,216. Thus, an incoming one-volt signal can be divided into more than 16 million steps on the Y-axis.
ADCs which offer high sample rates and high amplitude axis resolution is optimal for dynamic signal analysis such as shock and vibration. Low sample rates and high amplitude axis resolution are optimal for thermocouples and other measurands which have a wide amplitude range but which do not change state rapidly.
ADCs which provide anti-aliasing filtering (AAF) is highly desirable in all applications involving dynamic measurements because they prevent measuring errors caused by sampling a signal at a too low rate. This aliasing is when a false signal is created by sampling too infrequently for a fast-changing signal.
If the sampling is not fine enough, the retrieved signal can be very different from the real one.
Image courtesy of WikiCommons
Once converted to digital, our signals (aka measurands) are processed by the computer subsystem in several ways. First and foremost they can be displayed to the test operator on the system’s screen for visual inspection and review. Most DAQ systems show the data in several popular formats, including a time-history aka “strip chart” (Y/T) display, as well as a numerical display. But other display types are available from many systems on the market today, including bar graphs, X-Y graphs, and more.
Today’s data acquisition systems typically utilize a solid-state hard disk drive (SSD or HDD) to stream data from the ADC subsystem to permanent storage. Writing the data to disk also allows it to be analyzed after the test is completed.
Most DAQ systems allow the data to be exported to different file formats for analysis using third-party software tools. Common data formats include CSV (Comma Separated Values), UNV (Universal File Format), and more.
DewesoftX data acquisition software which is included for free with our modular data acquisition systems is able to export data to both formats plus many others. See the complete list of supported file export formats.
Removable SSD data storage on our rugged SBOX processing computer and data loggers
One of the most critical functions of any DAQ system is the ability to visualize the data in real-time during data storage. Systems typically employ an integrated or separate flat-screen display, which can be configured in a variety of visual formats.
Waveform data can almost always be displayed as Y/T waveforms against a graph or grid, and in numeric form. But other graphical conventions can be employed in addition, such as bar graph meters, FFT (Fast Fourier Transform) frequency/magnitude graphs, and more.
Today’s most flexible DAQ systems allow the user to configure one or more displays freely using built-in graphical widgets in an easy manner. DewesoftX software offers several high quality built-in visual instruments:
All visual instrument offers different customization options with real-time visual feedback.
Typical display from a DewesoftX DAQ software running on any Dewesoft's data acquisition system, showing measured data in a variety of user-selectable graphs and visual “widgets”
Data acquisition systems provide an important visual reference to the state of the test in real-time. But after it has been stored in the DAQ system, the data can also be analyzed using tools either built into the DAQ system or third-party data analysis software.
As mentioned earlier, nearly every DAQ system on the market today has several built-in data export filters that convert the system’s proprietary data format to third-party data formats for offline analysis.
Dewesoft data acquisition systems provide a wide array of data analysis features inside Dewesoft X data acquisition software
Today's data acquisition systems come in many forms and flavors. From modular data acquisition systems to all-in-one DAQs. There are many to choose from that suit certain applications. To learn about different types of data acquisition systems please check the following article - Types of data acquisition systems.
Data acquisition systems are sold by a variety of companies and are available with a broad range of capabilities and specifications, thus the prices can vary significantly. It is useful to provide general pricing for these various levels of DAQ systems, using the price-per-channel model. Estimated prices are given in USD (US dollars):
Do-it-yourself DAQ systems are impossible to estimate because they cover a wide span from a few channels to a system that took 10 man-years to develop and/or which involve hundreds or even thousands of channels.
