May 22, 2018 Uncategorized No Comments

PdM Strategies to Detect, Analyze and Correct

Efficient production and manufacturing depend on the harmonious contributions of multiple parties and machines. All must work together to ensure that production schedules are kept and the quality of end products meet client demands. Despite operating within an industry where machine disruptions and delays significantly affect the bottom line, many companies continue to employ reactive strategies when managing production roadblocks. It’s time to implement predictive maintenance strategies in the manufacturing process to save both time and money.

Detect

Early detection of potential machine failures is the goal for all predictive maintenance programs. Early detection allows maintenance teams to plan for repairs or change outs during planned outages to minimize downtime.

Vibration and ultrasound tools are at the heart of a good predictive maintenance program. Vibration will detect machine issues like unbalance, misalignment, looseness, etc. early enough that they can be corrected to prevent bearing failure. Once the bearing starts to fail, ultrasound will find those faults first – an early warning that can be monitored. Bearing faults will show up in the vibration spectrum as the fault progresses.

Vibration and ultrasound measurements can be trended to monitor machine health. These technologies work together to help us understand what actions can be taken to corre

ct the problem early. They can also predict how much time we have to plan maintenance.

Trending is a very important part of the predictive process. Velocity is measured in inches per second (ips) or millimeters per second (mm/s). These measurements are compared to ISO 10816 Alert and Alarm levels. Exceeding the Alert levels is our first indication that we have a problem.

Ultrasound is measured in decibels (dB). Ultrasound standards developed by NASA can also be compared to the ultrasound trend. Compared to the collected baseline, an increase of 8 dB is an indication that a bearing is under lubricated. Lubricating the bearing will lower the ultrasound level back to an acceptable level and a new baseline can be established. An increase of 12 dB means that the bearing has moved into late-stage bearing failure.

Using vibration and ultrasound trends together gives us a much better picture of machine health.

Analyze

We watch trends until we get an alert or alarm. Once the measurement has exceeded our ISO acceptable vibration levels, it’s time to analyze the data. This can be done by trained analysts, online analysis tools or remote third-party professionals.

Once we know something is wrong, we need to find out what it is. Analysis can tell us how to mitigate the problem so we can catch it early. Faults like unbalance, misalignment and looseness cause the majority of machine problems, leading to bearing failure, and early intervention can prevent machine failure.
Once we have identified a problem, we can analyze the data for comparative analysis of like machines. The results can be applied to the other motors, pumps, fans, compressors, etc. to ensure that the same fault isn’t present. Many problems we encounter with our machines are a result of the installation process, so if all of the pumps and motors in the pump room were installed at the same time, it makes sense to another inspection to rule out the fault.

A complete Failure Mode and Effects Analysis (FMEA) should be performed to determine root cause and eliminate and prevent this fault from occurring in the future.

Correct

Detection and analysis are a big part of the process, but correction is where the rubber meets the road. Determining when to intervene can be difficult. Here are some questions to ask:

  • Can or should production be interrupted to perform the repairs?
  • What is the cost of repair and downtime if we run it to failure?
  • When is our next maintenance outage and will it last that long?

Once a machine fault is found, it needs to be documented. Planning and scheduling need to happen. Parts and the proper tools need to be part of the planning and scheduling process, as well as the scheduling of the technicians and millwrights with the training and knowledge to do it right the first time.

Conclusion

Detecting, analyzing and correcting maintenance problems early can reduce maintenance costs by as much as 50% and increase uptime by as much as 30%. Proper training to use and apply predictive maintenance tools is essential to your success. Not only training to use the tools, but certification training to understand and analyze the data you collect from your machines is also critical.

Predictive maintenance solutions are becoming less expensive, simpler to use and ubiquitous. Is this your year to start a PdM program?

About GTI Predictive Technology

Paul Berberian is a Condition Maintenance Specialist at GTI Predictive Technology. At GTI Predictive Technology, it is our mission to provide the best predictive tools on a single platform and services to monitor nearly any asset. We strive to bring our customers the portability, connectivity, and affordability offered by the latest available technologies. Our product family combines vibration data collection and analysis with balancing, shaft alignment, thermography, and ultrasound into an affordable and completely scalable solution on one simple to use the platform. GTI’s predictive technology apps feature many additions that have come directly from customers.