IIOT Monitoring: Predicting disruptive failure in a Paper Mill

​Paper mills are looking for methods to reduce operating costs, minimize maintenance activities, while optimizing plant availability and uptime.  The key to this process are techniques such as predictive and preventative maintenance using vibration monitoring as a key indicator of machine health.
 
Vibration monitoring and analysis of paper mill processes have been in vogue for number of years. If a roll bearing fails when a paper machine is operating at high speed, extensive machine damage and the loss of substantial amounts of paper can result. In addition, production stops until repairs are made and the machine is restarted. In the past, mills relied on periodic “walk-arounds” by technicians using portable data collectors to measure vibration levels at various points. However, if a bearing failed between data collection rounds (and a serious fault can develop in as little as eight hours), the result could be very costly. Also, mill safety supervisors favored finding a means of remote monitoring of vibration data so that data collection technicians would not have to go near the operating machines
 
This was the advent of vibration monitoring and equipment that once installed can be hard wired in to a control SCADA and HMI system or database to visualize the vibration (time series and frequency data). Plant maintenance technicians could then review this maintenance data and then analyze the vibration data to be able to identify potential failures. One of the major challenges in this model was the amount of data that was collected, and the effort needed to find the “needle in the haystack”. The analysis and diagnostic procedure was very manual, and this involved a significant investment of time by the maintenance and engineering teams in being able to identify failure. A fault in one of the rollers in a paper mill would result in a disruptive failure resulting in a complete shutdown of the mill. The other issue with such a hard-wired monitoring system was the sheer installation cost and the technical challenges involved in  the installation.
 
One example of this was at a paper mill – that produced 300K Tm/year of Kraft paper; they followed the Kraft process for paper manufacturing, which involved the following elements: Impregnation, Cooking, Recovery, Blowing, Screening, Washing and Bleaching. This was then followed by the “Press Area” and the “Dry Area”.  The following shows a schematic of a typical paper mill
 

---

​The paper machine worked with a velocity of 750 m/min and with a weight of paper between 115 and 300 g/m; electric drives were AC with a variable speed, frequency variator.
 
The plant had been using vibration monitoring as a fundamental tool for equipment monitoring for more than 12 years. Most of the initial installation relied on analog devices and equipment that was installed in the 90’s, but in 2008, they decided to revamp their monitoring systems using a digital interface.
 
The initial installations were on the press areas and drying areas as shown in the diagram below
 

The press area of the paper machine has the harshest environment for the monitoring system due to the water pressure used for cleaning and the presence of chemical products like caustic soda and ClH in dissolution. The dryer area is a relatively safer place for the sensors but there are a number of sensors that need installation. Due to the sheer number of rollers that could breakdown in a commercial paper mill, there was a need for almost 600 – 700 sensors in these areas alone for one paper assembly line. A possible breakdown could include a bearing failure of anyone of the rollers and this would result in total unavailability of plant, downtime, lost production and significant economic impact.
 
 
But the initial installations were hard wired, laborious and took a lot of time – the installation took almost 21 days and involved a complete shut down by the plant for safety reasons. Conduit pipes and Cables had to be installed, right pipes for the cables and right layout of conduit needed to be installed to avoid difficulties in the normal operations of cleaning of the machine, or replacement of the floor cloths. About 20-30 boxes of sensor unions were then installed from which multiple cables were drawn to the final destination; namely the electrical boxes. Then the electrical boxes and associated wiring were installed for the entire plant. Additionally, the installation needed 5 monitors, with a total of 80 channels of measurement per box, a local system and database to store and visualize the data.
 
Today, advances in cloud, wireless and data analytics have enabled such installations to be completed in less than half the time. A sensor mesh in a plant allows Wifi enabled “Smart Sensors” to connect instantaneously. Installations today do not tap in to any hard-wired controllers, network boxes or cable boxes, the sensors connect directly to a gateway or router and then broadcast sensor data to the cloud. The additional advantage is that the data can be securely saved and visualized anywhere, you are not bound to the terminal in the paper plant. The mill’s chief vibration analyst can log onto the system from home, so he is on alert 24/7. If there’s an alarm, he can immediately access historical data on that point and analyze the vibration, looking at the trend lines to see exactly what happened and when. All the information needed to predict if and when a roller or bearing will fail is available, so maintenance supervisors can determine when to take action to avoid an unwanted machine shutdown.

---

Our experience in such installations suggests that it is still important to consider the following factors

• the easiest and most reliable way to install the monitoring system hardware;
• the most effective vibration measurements for detecting paper machine problems;
• the best setting for alarms to automatically detect machine problems
  
• appropriate responses to detected problems, including the most effective analysis methods and techniques;
  
•  documentation of vibration monitoring system events so as to best schedule repairs; and
• how to use documentation as a basis for improving effectiveness.

 
One of the very first monitoring points in a paper mill, revealed serious paper roll vibration, which was diagnosed as an improperly installed roller bearing. That catch prevented machine downtime and saved the mill nearly $40,000.
 
Talk to us about how Ennosys can help you digitize and transform your enterprise monitoring model for your paper mill!