Real-time monitoring of machine status enables prediction of future failures
The role of measurement hardware in measuring sensor change values is important.

Failure to address machine defects in a timely manner can lead to major accidents caused by sudden breakdowns. Therefore, regular preventive maintenance of industrial facilities is conducted to prevent casualties and property damage in industrial sites, but the costs are substantial. This is because significant waste occurs due to the replacement of not only defective parts but also healthy ones. Furthermore, since it is performed only on a regular basis, there are limitations in preventing sudden breakdowns. For these reasons, Prognostics and Health Management (PHM) technology is emerging as an alternative to regular preventive maintenance in almost all industrial sectors.

PHM technology monitors machine conditions in real time. It detects machine anomalies such as wear, misalignment, and contamination early and predicts future failures. For example, a vibration sensor can detect signs of failure about two months before it occurs. This allows appropriate measures to be taken in advance, reducing unnecessary maintenance costs and increasing stability and reliability.

Research into PHM technology was first initiated in the 1980s by the UK Civil Aviation Authority (CAA) to reduce the helicopter accident rate, which is 30 times higher than that of aircraft. In the 1990s, the CAA developed the Health & Usage Monitoring System (HUMS) to monitor the health of helicopters. When the system was installed and operated on actual helicopters, the accident rate was reduced to less than half of the previous rate. Since then, research into PHM technology began in earnest, and in Korea, the Korean PHM Society was launched last February and has already held two symposiums.

The core of PHM technology is the sensor. Each sensor installed in the machine detects changes in equipment vibration, speed, motor current, temperature, pressure, and the like. Supporting these sensors is the measurement hardware. The measurement hardware measures the changes detected by the sensors. To predict machine failures, one must know how to analyze and utilize the sensor outputs measured by the hardware.

When selecting measurement hardware, sensor characteristics such as frequency range, voltage range, and signal conditioning requirements must be considered. National Instruments (NI) has released a condition monitoring system measurement module capable of encompassing the characteristics of various sensors.

NI's NI 9232 is a general-purpose dynamic module that digitizes sensor outputs such as accelerometers, velocity, strain, speed sensors, and dynamic inputs. It features a 24-bit delta-sigma ADC with a resolution, a voltage range of ±30V, and a sampling rate of 102.4kHz. EPE driving is 4mA, 24V, and is ON/OFF configurable; the AC coupling cutoff is 0.1Hz and is also ON/OFF configurable. It has three channels and three 2-pin screw terminals.

By mounting multiple NI 9232 modules on NI's CMS-9068 device node, up to 24 channels can be obtained per chassis. There are 14 NI measurement modules for the CMS-9068, including the NI 9232 module. The NI 9207 module provides eight ±10V inputs and eight -21mA to 20mA inputs, and includes a 50/60Hz noise rejection option. The NI 9205 provides 16 differential inputs of ±10V. The NI 9208 provides 16 inputs of ±21mA.