This blog presents a brief and general introduction to data acquisition systems, also referred to as DAQ or DAS. These are instruments used to acquire signals from multiple sensors and transducers in order to store, analyze or act upon the acquired information. Some of the specific characteristics of the internal operation of DAQs have been discussed in previous blogs, or will be covered in other blogs to come.
Typical Configuration of DAQ Systems
DAQ systems can take multiple forms and have many levels of complexity, but they consist mainly of the same elements, usually arranged according to the basic configuration shown in the following figure:
- The sensors (or transducers) transform the information that is captured (sound, light, pressure, radio waves, and so on) into electrical signals.
- The analog front end further transforms and/or conditions the electrical signals from the sensors into other electrical signals that are adapted and optimized for proper acquisition by the DAQ system.
- The DAQ acquisition and processing section digitizes the electrical signals from the front end and processes them according to the specific needs of the application.
- A host PC stores, analyzes and/or further processes the stream of data generated by the DAQ system. A graphical user interface (GUI) lets the user interact with the DAQ system and its acquired data in a user-friendly manner.
- In user-programmable DAQ systems such as Nutaq’s µDigitizer (as well as our other larger acquisition systems), high-level design tools can be used on the host computer to completely reprogram the acquisition and processing functions performed by the DAQ system.
DAQ systems can be very complex, both in their configuration and in the processing they have to perform on the acquired data. A successful DAQ implementation relies on an optimized combination of high-performance, well-integrated hardware and sophisticated high-level software tools to develop and control the operation of the instrument.
The Analog Front End
The physical configuration of the analog front end is determined by the specific requirements of the application, and usually needs to be partially or totally custom designed. The type of sensors, their number, their frequency bandwidth and the electrical characteristics of their output signals will all contribute to the configuration and design of the analog front end.
The front end is used to apply various types of transformations to the electrical signals from the sensors, such as amplification (gain), filtering, bandwidth limiting, frequency translation, single-ended to differential, and so on. The front end is also often used as a bridge between the specific physical constraints of the sensors and the mechanical/electrical interface requirements of the DAQ system.
DAQ Acquisition and Processing
The acquisition and processing section is usually the most complex part of any DAQ system. Its typical internal components are shown below:
At minimum, the acquisition and processing section consists of a single “DAQ Block” (shown in blue) that digitizes and processes a certain number of signals (channels) from the analog front end. In more elaborate systems, multiple DAQ blocks can be combined to increase the number of supported channels.
Using multiple blocks requires that these blocks be synchronized with each other so that the data they generate share a common sampling clock and preserve the relative timing information between channels. This function is accomplished using a dedicated clock and synchronization module (shown in green) that generates multiple phase-coherent sampling clocks and synchronization pulses for each of the DAQ blocks.
The data generated at the output of the DAQ blocks is usually sent to a host computer over a high-speed communication link such as Ethernet, Gigabit Ethernet or PCIe (PCI Express). In systems having multiple DAQ blocks, a dedicated communication switch (shown in purple) handles the information traffic between the various DAQ blocks and the host computer. In certain configurations, where the host computer is embedded in the DAQ system itself, the communication traffic can instead be handled over an internal backplane data bus such as CompactPCI or VME.
Each DAQ block uses an analog-to-digital converter (ADC) to digitize the signals from the analog front end. The maximum number of channels per DAQ block (“M” on the diagram) is determined by the capabilities of the selected ADC hardware. The “M” channels of digital data generated by the ADC are usually processed by an FPGA, which executes an algorithm that is specific to the application.
The FPGA sometimes has to use large amounts of external memory to acquire or buffer the incoming or outgoing data, or to temporarily store information during the execution of the processing algorithm. The FPGA also sometimes uses external signals to control the operation of the analog front end from a distance.
The Host PC
The host PC serves as the control center of the DAQ system. It can be used to:
- control the DAQ
- provide a graphical user interface (GUI) for the user to operate the system
- store the acquired information on mass storage devices
- provide the tools to display and/or analyze the acquired data in real or non-real time
In advanced development platforms such as the ones provided by Nutaq, the host computer is also used to define and implement the algorithms performed by the DAQ system. Nutaq’s tools implement what is called the “Model-Based Design Kit”, or MBDK. Together with MathWorks® Simulink® and Xilinx® System Generator for DSP™, the MBDK creates a graphical and user-friendly framework that greatly simplifies the burden of developing and porting an algorithm to a dedicated hardware platform. Using these tools, it is possible to control the behaviour of the DAQ system without having to write a single line of C or VHDL code.
For More Information
If you’re interested in learning more about specific aspects of DAQ systems, please read our series of blogs on the subject at http://nutaq.com/application/multichannel-daq.