Data Acquisition Systems: Components, Types & Applications

TAKEAWAYS:

• A Data Acquisition System (DAS) is a hardware/software framework that measures, digitizes, and processes real-world signals (e.g., temperature, pressure, vibration) into usable data for analysis or control.
• Single-channel DAS consists of a sensor → signal conditioner → ADC → microcontroller → display/recorder, while multi-channel systems use multiplexers to handle multiple inputs sequentially.
• Key components include transducers (e.g., LVDT, thermocouples), signal conditioners (amplifiers, filters), ADCs (e.g., 8-bit to 24-bit), and data loggers (e.g., Arduino, LabVIEW).
• Applications span industrial automation (e.g., NTC’s power grid monitoring), medical devices (e.g., ECG machines), and IoT (e.g., smart agriculture sensors).
• Errors and calibration in DAS arise from sensor drift, ADC quantization, and noise; mitigation includes shielding, differential measurements, and periodic calibration.
• Exam focus: Block diagrams, LVDT operation, single-channel DAS workflows, and real-world traces (e.g., Daraz’s inventory tracking via RFID-DAS).

1. What is a Data Acquisition System (DAS)?

A Data Acquisition System (DAS) is an integrated system that:

1. Senses physical phenomena (e.g., temperature, displacement, voltage) via transducers/sensors.
2. Conditions the raw signal (amplification, filtering, linearization).
3. Converts analog signals to digital (ADC).
4. Processes data (microcontroller, FPGA, or PC).
5. Stores/Displays results (data loggers, SCADA, or cloud platforms).

Why is DAS critical? Without DAS, we couldn’t:

• Monitor Nepal’s NTC power grid in real-time (voltage/current fluctuations).
• Track Daraz’s warehouse inventory via RFID readers + DAS.
• Diagnose patient vitals in hospitals (ECG, blood pressure).

2. Core Components of a DAS

A DAS is built from five layers, visualized below:

A. Sensors/Transducers

Convert physical quantities into electrical signals (voltage, current). Examples:

B. Signal Conditioning

Raw sensor signals are often noisy, weak, or nonlinear. Conditioning includes:

1. Amplification: Boosts signal strength (e.g., instrumentation amplifier for LVDT).
2. Filtering: Removes noise (e.g., low-pass filter for ECG signals).
3. Linearization: Corrects nonlinear sensor outputs (e.g., thermocouple lookup tables).
4. Isolation: Protects from electrical interference (optocouplers, transformers).

Example: In a Khalti payment terminal, the magnetic stripe reader’s weak analog signal is amplified and filtered before ADC conversion to avoid misreads.

3. Single-Channel DAS: Step-by-Step Workflow

A single-channel DAS processes one signal at a time. Below is the trace for measuring temperature in a Daraz cold storage using an RTD:

sequenceDiagram
    participant RTD as "RTD (Pt-100)"
    participant Amp as "Signal Conditioner\n(Amplifier + Filter)"
    participant ADC as "12-bit ADC\n(e.g., MAX11100)"
    participant MCU as "Microcontroller\n(STM32)"
    participant SD as "SD Card Logger"

    RTD->>Amp: Output: 138.5Ω (25°C)
    Amp->>Amp: Amplify to 0–5V range
    Amp->>ADC: 2.5V (digital: 0x800)
    ADC->>MCU: 25.0°C (after linearization)
    MCU->>SD: Log timestamp + temp

Worked Example: RTD Temperature Measurement

1. Sensor: Pt-100 RTD at 25°C → Resistance = 100Ω + (0.385Ω/°C × 25) = 138.5Ω.
2. Excitation: 1mA current → Voltage drop = 138.5Ω × 1mA = 138.5mV.
3. Amplification: Gain = 10 → Output = 1.385V.
4. ADC Conversion: 12-bit ADC with 0–5V range → Resolution = 5V/4096 ≈ 1.22mV/LSB.
   • Digital output = 1.385V / 1.22mV ≈ 1133 (0x469).
5. Linearization: MCU converts digital code to temperature using: $$T=\frac{R-R_0}{\alpha }=\frac{138.5-100}{0.00385}\approx 25.0°C$$
6. Storage: MCU logs 2024-05-20 14:30: 25.0°C to SD card.

4. Multi-Channel DAS: Handling Multiple Signals

For multiple inputs (e.g., monitoring 10 sensors in a NTC substation), a multiplexer (MUX) switches signals to a single ADC.

Advantages of Multi-Channel DAS:

• Cost-effective: Single ADC for multiple sensors.
• Scalable: Add more channels via MUX expansion.
• Synchronized sampling: Critical for correlated data (e.g., ECG + blood pressure).

Disadvantage:

• Speed trade-off: Sampling rate per channel = Total rate / Number of channels.

5. Key Instruments in DAS

6. Errors and Calibration in DAS

Common Errors:

1. Sensor Errors:
   • Hysteresis: LVDT output lag due to mechanical friction.
   • Nonlinearity: Thermocouple output deviates from ideal.
2. ADC Errors:
   • Quantization Error: ±0.5 LSB (e.g., 12-bit ADC has ±0.024% error).
   • Offset Error: ADC reads 0V as 2 LSBs instead of 0.
3. Noise:
   • Electromagnetic Interference (EMI): Affects long cables (e.g., NTC power lines).
   • Thermal Noise: Random fluctuations in sensors (e.g., strain gauges).

Calibration Methods:

• Two-Point Calibration: Adjust gain/offset using known inputs (e.g., ice point and boiling point for thermocouples).
• Auto-Calibration: MCU runs self-tests at startup (e.g., measuring 0V and 5V reference).
• Periodic Checks: NTC calibrates power meters annually using standard resistors.

7. Applications of DAS

A. Industrial Automation (Nepal)

• NTC Power Grid Monitoring:
  • DAS Role: Measures voltage/current in substations via CTs (Current Transformers) + ADCs.
  • Example: A 16-channel DAS logs phase angles to detect faults in Kathmandu’s transmission lines.
  • IMAGE: current transformer (CT) labelled diagram | Primary/secondary windings, burden resistor.

B. Medical Devices

• ECG Machine:
  • DAS Role: Amplifies heart signals (1mV peak) → filters 50/60Hz noise → ADC → displays waveform.
  • Example: A 12-lead ECG uses 8-channel DAS with 24-bit ADCs for high fidelity.

C. IoT and Smart Cities

• Smart Agriculture (Nepal):
  • DAS Role: Soil moisture sensors (e.g., capacitive sensors) → WiFi module → cloud logging.
  • Example: Pathao’s electric scooters use DAS to monitor battery temperature/voltage for safety.

D. Financial Systems

• ATM Transaction Logging:
  • DAS Role: Records card swipe signals, PIN entry timing, and transaction data.
  • Example: Nabil Bank ATMs use DAS to detect fraud via abnormal signal patterns.

8. Comparison: Single-Channel vs. Multi-Channel DAS

9. Worked Example: LVDT in a Hydraulic Press

Scenario: A factory uses a hydraulic press with an LVDT to measure piston displacement. The LVDT outputs:

• 0V at 0mm displacement,
• +5V at +10mm,
• -5V at -10mm.

Given:

• LVDT sensitivity = 0.5V/mm.
• ADC range = 0–10V, 10-bit resolution.

Steps:

1. Displacement = 6mm → LVDT output = 6mm × 0.5V/mm = 3V.
2. ADC Conversion:
   • Resolution = 10V / 1024 ≈ 9.77mV/LSB.
   • Digital code = 3V / 9.77mV ≈ 307 (0x133).
3. MCU Calculation: $$\text{Displacement}=\frac{\text{ADC code}-\text{Offset}}{256}\times 10\text{mm}$$ (Offset = 512 for ±10mm range). $$=\frac{307-512}{256}\times 10=-7.96\text{mm}(\text{Error due to quantization})$$

Correction: Use 12-bit ADC for ±0.2mm error.

10. Exam Tip: How to Score Full Marks

1. Block Diagrams:

   • Always draw single-channel DAS with labeled blocks (sensor → conditioner → ADC → MCU → display).
   • For multi-channel, show the MUX and sampling sequence.

2. LVDT Operation:

   • Explain differential output (two secondary coils 180° out of phase).
   • Derive displacement from voltage ratio: $$V_{out}=k\cdot x\text{(where }k=\text{sensitivity)}$$

3. ADC Calculations:

   • Memorize: $$\text{Resolution (LSB)}=\frac{V_{ref}}{2^n}$$
   • Example: For 5V, 8-bit ADC → LSB = 19.53mV.

4. Real-World Traces:

   • Link to NTC power monitoring or Daraz inventory systems.
   • Example: "A 16-channel DAS in a NTC substation samples 1000Hz per channel, logging phase angles to detect faults."

5. Error Analysis:

   • Mention quantization error, sensor nonlinearity, and EMI shielding.
   • Example: "A thermocouple’s 10°C error at 100°C is due to nonlinearity; use a lookup table in the MCU."

In the Real World

1. eSewa’s Payment Terminals:

   • DAS Role: Magnetic stripe readers use Hall-effect sensors → signal conditioner → ADC → encrypts card data.
   • Why? Ensures clean analog-to-digital conversion to avoid transaction errors.

2. NTC’s Smart Grid:

   • DAS Role: Phasor Measurement Units (PMUs) sample voltage/current at 30–60Hz using 16-bit ADCs to detect grid instability.
   • Example: During a 2022 Nepal blackout, PMU-DAS identified a transformer fault in Chitwan.

3. Pathao’s Ride Safety:

   • DAS Role: Accelerometers in scooters measure vibration and tilt (via ADC) to detect reckless driving.
   • How? A 3-axis accelerometer outputs ±2g → conditioned → 12-bit ADC → cloud alert if tilt > 30°.

Key Equations to Remember

Common Pitfalls in Exams

• Forgetting the MUX in multi-channel DAS → Lose marks on sampling rate explanation.
• Ignoring signal conditioning → ADC will saturate if input exceeds range.
• Mixing single-ended and differential inputs → LVDT requires differential measurement.
• Not calibrating → Assume sensors are ideal unless stated.