Data Acquisition Systems
A data acquisition system turns physical quantities into timestamped digital data, and its timing and integrity determine what control and analysis can achieve.
The Bridge from Physics to Data
A data acquisition (DAQ) system captures physical quantities and delivers them as digital values a computer can use. It spans the chain from sensor to stored or streamed numbers: signal conditioning, filtering, conversion, timestamping, and transfer. The quality of everything downstream, from control loops to offline analysis, is capped by the fidelity and timing of this front end.
The Signal Chain
- Sensor and transducer converting a physical quantity to an electrical signal
- Signal conditioning: amplification, isolation, and level shifting
- Anti-alias filtering before sampling
- Analog-to-digital conversion
- Timestamping and buffering
- Transfer to the processing or storage system
Timing and Synchronization
For multi-channel measurements, when each channel is sampled relative to the others matters as much as the values. Simultaneous sampling uses one converter per channel or a sample-and-hold on each so all channels capture the same instant, which is essential when computing relationships between signals. Multiplexed systems share one converter and stagger channels in time, adding a known skew that must be accounted for. Precise time synchronization across distributed units keeps a common timebase.
Throughput and Buffering
High channel counts and high sample rates produce large data streams. DAQ systems use direct memory access and buffering to move data without burdening the processor, and they must guarantee no samples are lost. Overflow of a buffer is a data-integrity failure, so buffer sizing and transfer bandwidth are part of the timing budget.
Integrity and Provenance
Trustworthy data carries its own metadata: calibration, timestamp source, channel configuration, and any gaps. In experimental and control contexts, such as fusion diagnostics, thousands of channels are acquired at once and later reconstructed into physical quantities, so every sample must be attributable to a known time and calibration. A DAQ system's job is not only to capture data but to capture it in a form whose accuracy and timing can be defended.