Technical Article

Different Techniques for Condition Monitoring

August 29, 2020 by Muhammad Asim Niazi

Condition monitoring focuses on the physical parameters like temperature and vibration, rather than the product parameters.

Condition monitoring is a technique that involves measuring the condition of the equipment. These physical parameters indicate the component's present trend, and this trend is used to predict when its performance will go in a failure condition.

It focuses on the analysis of the degradation or the rate of failure rather than general-purpose monitoring. The data is collected with the sensors and fed to the appropriate system like the predictive maintenance system for processing and utilization.

 

What Techniques are Used for Condition Monitoring?

The condition monitoring system can be implemented through many techniques. All techniques are important and offer proven results. However, the selection of a specific method depends upon the application and use of the component.

These techniques can be used as a standalone or combination of different techniques including: 

  • Oil analysis
  • Vibration analysis
  • Motor current analysis
  • Thermography
  • Ultrasonic monitoring

 

A variety of condition monitoring technologies help asset specialists understand asset health. Image courtesy of National Instruments Corp.

 

Oil Analysis

By analyzing oil and studying any changes in its properties, we can have insight into the component's condition. The condition of the oil is affected by the poor function of the equipment. The poor function changes the characteristics of the oil, and it also adds contaminants to it. 

 

Oil analysis being shown. Image courtesy of the Maintenance Institute Africa. 

 

By studying contaminants, we can have an idea about the source of these contaminants enabling us to perform corrective actions before going into the breakdown.

This technique is useful for systems like gearboxes, transmission systems, combustion engines for power requirements, hydraulic systems, and compressors. 

 

Vibration Analysis

Every component that is meant for any movement faces some vibration. However, excessive vibration is considered bad or abnormal. Abnormal vibration is an essential contributor to a problem in the systems.

By performing continuous monitoring of vibration in the mechanical & rotating components, we can detect fault before it happens. This is because the change in vibration indicates a change in equipment behavior and this behavior change is due to some abnormalities.

 

Assessing the vibration analysis of a system. Image courtesy of Lakeside Process Controls.

 

Some of the best applications of vibration analysis to detect the condition can be:

  • Bearing
  • Gearbox 
  • Shafts
  • Motor 
  • Blower fans
  • Piston engines 
  • Compressors 
  • Pumps

 

Motor Current Analysis

This technique is used for the induction motors. By analyzing a running motor's behavior, we can successfully predict the breakdown or a faulty condition of the motor. In this technique, the current is observed for a running motor. Any change in the trend of the current is considered to be an abnormal condition.

The advantage of the technique is that the sensors are installed in the electrical cabinets rather than at the motor's location. With this technique, no additional installations are required.

The data is then fed to the software. The algorithm of the software calculates the performance of the motor. Any abnormalities are detected by the software and mentioned in the result. The result is then used and evaluated by the concerned persons to take notice and plan action accordingly.

The type of problems that they can detect, but not limited to the following:

  • Loose windings
  • Electrical supply unbalance
  • Eccentric rotors
  • Problems in rotor bars like degradation, damage
  • Problems with the bearings

 

Thermography

In thermography, the temperature or heat pattern is recorded of the system under consideration. Since every component emits radiation even under normal working conditions, we can track this thermal activity to detect any problem.

Any malfunction in a system increases thermal emission, greater than the normal value. If there are some abnormal working conditions in a machine like high load due to jamming in a motor, aging of components, then it also results in higher than the normal temperature values.

Suppose we record these normal values and compare these values to the measured values taken at specific time intervals. In that case, we can detect any faulty condition of the equipment or if there is a deviation from the normal curve.

One of the main advantages of this technique is that we can conduct this activity safely from the equipment under inspection. This means that dangerous equipment can be inspected without risking life. Some of the typical applications include but not limited to are:

  • Electrical panels
  • Electrical wires
  • Substations and Transformer
  • Motor and their associated components like bearings, housing, and internal winding
  • Troubleshooting piping related issues
  • HVAC system, duct leakage

 

Ultrasonic Monitoring

Ultrasonic frequency belongs to the spectrum of frequencies beyond the range of human hearing capabilities and typically lies in the range of 20 kHz to 100 kHz.

Ultrasonic frequencies are also emitted by mechanical equipment, due to some electrical emissions resulting from leakage in a closed system like in a piping network. By measuring and monitoring these ultrasonic frequencies, we can have insight into the system's condition. Modern instruments not only measure these frequencies but also store data. It also enables us to analyze data through an algorithm.

Why use ultrasonic frequencies in the presence of other techniques? Ultrasonic frequencies are a short wave in nature, and due to this property:

  • The condition of pinpoint location can be found out
  • Works well in a noisy environment
  • An early indication of mechanical equipment

 

Difference Between Predictive and Condition-Based Maintenance

Predicative and condition-based maintenance are mistakenly considered to be the same. This is because these both involve analyzing the system and take action before some fault occurs. However, this is not the case. These two are necessary for each other to be successfully implemented.

Condition-based is data and predictive based is an algorithm. Condition-based takes the necessary inputs while the predictive maintenance plans the maintenance taking feedback from condition-based.

 

Solution or No Solution

While condition-based maintenance is responsible for monitoring the different components and gives the actual condition, it does not give a solution based on the condition of the equipment. On the other hand, predictive maintenance suggests solutions, so that maintenance is carried out before the problem arises.

 

Time Dependency 

Condition-based maintenance is dependent upon the condition of a system. If the condition is satisfactory, no matter how many hours it has been running, then maintenance activity is not planned.

Predictive maintenance is planned at a particular time, depending upon the feedback received from different scenarios and devices. This predicts the possible failure and initiates the maintenance before it happens.