Sensors for Condition Monitoring

Overview

Condition monitoring systems monitor equipment degradation. Typically, an equipment criticality and reliability assessment is performed to identify important equipment and common failure modes. With the selected equipment and failure modes in hand, you can identify sensors to monitor specific components of the equipment for degradation and the initiation of identified failure modes.

Measurement Technologies and Sensors for Condition Monitoring Applications

Many sensors detect changes in equipment components. Vibration sensors are the most commonly used sensors when monitoring rotating machinery. They are often credited for sensing the mechanical degradation of equipment components two months or so in advance of a failure, as shown in Figure 1.

Figure 1: Asset degradation can be detected in advance by using sensors.

Other sensing technologies used in equipment condition monitoring as shown in Figure 2 include motor current, oil analysis, thermography, dynamic pressures, and operating state sensors such as speed.

Figure 2. A variety of condition monitoring technologies help asset specialists understand asset health.

Table 1 lists the types of sensors and sample vendors used in condition monitoring.

MeasurementSensorFrequency RangePossible Signal Conditioning NeedsVendors
VibrationAccelerometer>100 Hz

IEPE

AC/DC coupling

±24 V input or AC couple

Anti-alias filter

IMI Sensors

Connection Technology Corporation

Endevco/Wilcoxon

VibrationVelocity>20 Hz to <2 kHz

IEPE

AC/DC coupling

±24 V input or AC couple

Anti-alias filter

IMI Sensors

Connection Technology Corporation

Endevco/Wilcoxon

VibrationProximity Probe (Displacement)<300 Hz

Modulator/demodulator Anti-alias filter

±30 V input range

Connection Technology Corporation
SpeedProximity Probe<300 Hz

Modulator/demodulator Anti-alias filter

±30 V input range

Connection Technology Corporation
SpeedMagnetic Zero SpeedUp to 15 kHz

24 V DC power

±20 V

Honeywell

SPECTEC

Motor Current

Current Shunt

Current Clamp

Up to 50 kHz±333 mV or ±5 VMagnelab
Temperature

RTD

Thermocouple

Up to 10 HzNoise rejection, excitation, cold-junction compensationNI
TemperatureInfrared CameraMultiple frames per secGigE Vision over EtherNet connectionFLIR Systems
PressureDynamic Pressure>100 Hz

AC/DC coupling

IEPE (some models)

±24 V or AC coupling

Anti-aliasing filter

Endevco

PCB

Kulite

Kistler

Oil Quality

Oil Particulate

Viscosity

Contamination

Particulates

Up to 10 Hz

mA current input

±10 V input

50/60 Hz noise rejection

Kittiwake

Honeywell

HYDAC

Poseidon Systems

High-Frequency “Noise”Ultrasonic>20 kHz

AC/DC coupling

±24 V input range

Anti-aliasing filter

UE Systems

 

Table 1: There are a variety of measurements, signal condition needs, and vendors for condition monitoring sensors.

Figure 3 shows a typical sensor and a sample failure mode of an asset the sensor monitors.

Figure 3: Typical sensors and failure modes that can be detected.

Each sensor offers the ability to monitor the degradation of mechanical and electrical components within rotating machine equipment.

  1. Vibration sensors are used to detect roller bearing wear, gearbox wear, shaft misalignment, unbalance, and mechanical looseness.
  2. Speed sensors work with vibration sensors to correlate vibrations to rotating speed and shaft angular position.
  3. Motor current sensors are commonly placed at the motor control center. They can detect eccentric rotors, loose windings, rotor bar degradation, and electrical supply unbalance.
  4. Dynamic pressure sensors are used for combustion dynamics, flow turbulence, and cavitations.
  5. Temperature sensors are typically used to detect heat caused by friction. They often accompany vibration sensors to collaborate vibration-detected degradation.
  6. Thermal imaging detects hundreds of temperatures within the camera’s field of view.
  7. Ultrasonic sensors can detect electrical problems including corona, arcing, and tracking. They can also be used to detect early signs of roller bearing wear.
  8. Oil sensors can detect wear debris from bearings and gears. They also can detect contaminants in the oil that reduce the lubrication ability of the oil.

Once you have identified equipment, failure modes, and sensors, the next step is to choose measurement hardware to digitize the sensors. Your choice of measurement hardware depends on your sensor characteristics including frequency range, voltage range, and signal conditioning needs.

With respect to frequency range, you can choose from two types of sensors: static and dynamic. Dynamic sensors measure frequency ranges well above 2 Hz. For example, a typical accelerometer has a frequency range of 1 Hz or 2 Hz to 15 kHz or 20 kHz. The ultrasonic sensor frequency range extends to 100 kHz or more, with a heterodyne output ranging from a few hundred hertz to 40 kHz. On the contrary, static sensors convert the monitored physical property into an electrical signal at rates of 1 Hz to 10 Hz or 20 Hz.

The other two sensor characteristics are voltage range and signal conditioning needs. Some sensors produce a voltage as high as 30 V, while others might produce only a few millivolts. Some sensors need sensor power such as IEPE power, while others might need cold-junction compensation. Table 1 lists the common signal conditioning needs for each measurement and sensor.

NI provides a wide range of hardware to acquire data from these sensors as well as the software to process and analyze the data. For deployment-ready software that connects to these sensors, please see the link below.