US What a data acquisition system does
A data acquisition system (DAQ system, sometimes DAS) is the hardware and software that measures physical signals and records them as digital data. Data acquisition is the process; the system is the chain of parts that makes it happen, from the sensor on the machine to the file on the disk.
A DAQ system answers a simple question many times a second: what is this signal doing now? The answers become a time-stamped record of temperatures, pressures, strains, voltages, currents, speeds or switch states that engineers can plot, analyze, alarm on or feed into a control loop.
Unlike a chart recorder or a handheld meter, a DAQ system digitises the signal at the point of measurement and keeps it in a form software can process. That is what makes it the base of test stands, condition monitoring, machine control and research rigs.
The parts of a DAQ system
Every data acquisition system, from a single USB device to a multi-rack test cell, uses the same chain of parts.
-
//
Sensors and transducers. Thermocouples, RTDs, strain gauges, load cells, accelerometers, pressure transmitters and limit switches turn a physical quantity into an electrical signal.
-
//
Signal conditioning. Amplifies small signals, filters noise, isolates the input, powers or excites the sensor and scales the signal to the converter’s range.
-
//
Analogue-to-digital converter. The ADC samples the conditioned signal at a fixed rate and turns each sample into a number. Digital inputs, counters and encoder inputs sit alongside it for on/off and pulse signals.
-
//
Computer and interface. The DAQ hardware connects to a computer over USB, PCIe, a stacking bus such as PC/104, or Ethernet, and the computer timestamps and stores the data.
-
//
Software. Drivers and acquisition software set up channels, ranges and rates, then display, log, analyze and export the data or pass it to a control system or historian.
Sample rate and resolution
Two specifications decide whether a DAQ system sees the signal properly. Sample rate is how many readings per second each channel takes. By the Nyquist theorem it must be more than twice the highest frequency in the signal, and in practice engineers sample five to ten times faster to capture the waveform shape. Temperature may need one sample a second, while vibration analysis may need tens of thousands.
Resolution is how finely the ADC divides its input range. A 16-bit converter splits its range into 65,536 steps, so on a ±10 V range each step is about 0.3 mV. 24-bit converters, usually sigma-delta designs, suit vibration, acoustic and bridge measurements with a wide dynamic range. Resolution only helps if the signal fills the range, which is a job for signal conditioning.
Check also whether channels are sampled simultaneously or multiplexed through one ADC. Multiplexed inputs are cheaper but take readings a few microseconds apart, which matters when comparing the phase of fast signals.
Types of data acquisition system and their interfaces
DAQ systems are usually grouped by how they connect to the computer, which also sets where they can be installed.
| Interface | Form | Strengths | Watch out for | Typical use |
|---|---|---|---|---|
| USB | External box on a laptop or PC | Easy to move between computers; bus power on small devices | Cable length and latency; USB disconnects on vibrating machines | Bench test, portable logging |
| PCIe and Mini PCIe | Card inside the computer | High sample rates, low latency, tight timing with the host | Needs a free slot; channels wired into the computer | Test stands, machine control, high-speed capture |
| PC/104 and PCI/104-Express | Stacking module on an embedded board | Compact and rugged; screwed-down stack resists vibration | Limited stack height and power | Vehicles, rail, defense, embedded instruments |
| Ethernet | Networked chassis or remote I/O module | Long cable runs; sensors measured near the machine | Network timing and synchronization must be planned | Distributed plant monitoring, large structures |
| Stand-alone logger | Self-contained recorder with its own storage | Runs without a PC for days or weeks | Usually slower; analysis happens later | Field surveys, unattended monitoring |
Many industrial projects combine several types: a plug-in card for fast local signals, plus networked modules for slower readings spread across a site. When the inputs are spread out and mainly feed a control system, the arrangement is closer to remote I/O than a test-and-measurement DAQ.
Where data acquisition systems are used
Data acquisition sits behind most measurement-driven work in industry and research.
-
//
Test and validation. Engine, gearbox, battery and structural test rigs record hundreds of channels during each run.
-
//
Condition monitoring. Vibration and temperature readings on motors, pumps and fans reveal wear before a breakdown.
-
//
Process and environmental monitoring. 4–20 mA transmitters for pressure, flow and level are logged alongside control data.
-
//
Vehicles and rail. On-board systems log CAN data, strain, acceleration and position over a journey.
Choosing a DAQ system for an industrial computer
Start from the signals, not the hardware: count the channels of each type, note each one’s range and the bandwidth it needs, then add the environment, which sets the form factor and the isolation needed.
Our data acquisition and I/O modules are mainly board-level designs for embedded computers. They include analog I/O modules, mostly 16-bit and some with 4–20 mA outputs; digital I/O modules with optoisolated inputs and relay outputs; and isolated CAN interfaces. Most are PC/104, PCI/104-Express or Mini PCIe cards, and the PC/104 I/O modules stack directly on a PC/104 board for vehicle and rail systems.
Quick answers on data acquisition systems
Data acquisition systems: common questions
What does DAQ stand for?
Data acquisition. A DAQ device is the hardware that digitises signals, and a DAQ system is the full chain of sensors, conditioning, hardware, computer and software.
What are the three types of data acquisition system?
Usually PC-based, networked and stand-alone. PC-based systems plug into or connect directly to a computer, networked systems put the I/O near the sensors on Ethernet, and stand-alone loggers record on their own and are read later.
What is the difference between a DAQ system and a data logger?
Speed and independence. A data logger records a few slow channels on its own for long periods. A DAQ system usually works with a computer, samples faster and analyses the data as it arrives.
What software is used for data acquisition?
The vendor’s software or your own code. Most DAQ hardware ships with configuration and logging software plus drivers and libraries for C, C# or Python, and many SCADA and historian packages can read DAQ data directly.
What is an example of a data acquisition system?
A motor test stand. Thermocouples, a torque transducer, current clamps and an encoder feed a multifunction DAQ card in an industrial PC, which logs every run for comparison.
We fit data acquisition and I/O modules into industrial and embedded computers as standard, install the drivers, and check each channel before the system ships.