Barry Weller of Mitsubishi Electric discusses why data logging has become an essential feature of the modern production line and why it is often best performed within a PLC platform
These days it’s become almost a cliché to say you can’t manage what you don’t measure, yet it’s never been more important. As a result, data logging has quickly become an important feature of plants aspiring to world class manufacturing status.
It is perhaps strange that data loggers formed a critical part of test and measurement applications and the process arena for so long without impacting on discrete manufacturing production lines. But sensor derived information has always been acted upon in real time, so production line availability was seen as a black and white issue, but times have changed. Increased international competition means line efficiency is ever more important, while increasing litigation has placed more emphasis on traceability.
Production efficiency looks at many different aspects of a machine, line or process and rolls them all up into concepts such as overall equipment effectiveness (OEE). But behind it is the need to minimise downtime, increase quality, reduce scrap or rework and maximise availability. Along with other key performance indicators, OEE scores give an idea of how well a plant is running at any given moment, while providing a measure of the impact of changes in line operation over time.
With OEE, we need real time production information and we need to observe trends – which is where data logging on the plant floor comes in. Only with improved, time-stamped data, continuously collected, can production management be statistically analysed so that machine availability and product quality can be improved. On a single machine, bottlenecks or mechanical problems can be quickly identified and rectified. Logging and monitoring fault codes and error messages can highlight recurring problems and enable appropriate action to be taken. Trends in production tolerances can be monitored and any corrections made before they impact detrimentally on product quality.
While OEE brings a statistical focus to product quality, customers also need assured quality of the supply chain. If somewhere along the supply chain a fault or a problem with a product is discovered, the origin of that fault should be fully traceable. Such traceability has always been demanded in pharmaceutical and food processing, and increased litigation across all market sectors has seen a growing emphasis on track and trace. Building traceability into production lines means that faults can be identified and rectified long before they reach the customer and also that manufacturers can provide documentary evidence to show that any faults did not originate with them. Achieving this level of traceability depends upon the collection and recording of time-stamped data throughout production.
Beyond machine availability and quality, the total cost of production of a given part, also builds in the cost of energy to manufacture that part and looks at energy efficiency as a key performance indicator. By monitoring energy usage effectively, manufacturers can begin to optimise processes, manage peaks in demand and reduce cost per part. In addition, as climate change legislation starts to take affect, the need to monitor and record energy consumption grows.
Data loggers can collect and record energy consumption data, highlighting demand and illustrating the impact of any measures taken to reduce the energy usage.
Collecting and recording data
Given this requirement to collect and record data, there is the question of how this is best performed, and where. There are many data loggers and data acquisition systems available, all with their advantages and disadvantages depending on the data that needs to be collected. In many PLC-based automation processes, the ability to collect data directly within the PLC and automatically transfer that data in simple formats that make it easy to analyse provides the most suitable, cost-effective solution.
The PLC-based solution is attractive because the architecture for collecting the data is already in place. The PLC is already linked to sensors, actuators, drives, network components and other controllers, reading their inputs and acting upon them. There will inevitably be other data sources to monitor, but subsequent integration is simplified.
Automation vendors have long offered add-on data loggers for PLCs. Most recently, though, demand for a new breed of machine control PLCs has come with a corresponding demand for a new type of data logging solution. Mitsubishi Electric addressed the former with the L Series PLC, and the latter by integrating a full featured data logger within it, eliminating the need and cost of an add-on module.
The L Series PLC provides the functions and capabilities required for automated plants in a compact, rack-free package. Bridging the gap between the FX micro PLC and the iQ platform PAC, the new modular controller puts much of the power of a Q series CPU into a small PLC. The built-in data logging function provides a simple means for OEE monitoring, energy monitoring, sequencing and production traceability, with the facility to store this data to SD memory card using the integral memory card slot. This slot can also be used to back up or load the CPU programs and parameters.
The built-in data logger supports asynchronous scan-independent sampling as fast as 1ms. Separating its operation from the cyclic scan of the PLC provides far greater flexibility and sample frequency. Configuration of the data logger is made with parameterisation rather than programming, saving setup time. This also allows the reports to be quickly reconfigured to capture the data that is needed, and analysis of the logged data is available in Excel format or in the configuration tool. There is also an auto-logging function which allows a ‘setting file’ to be emailed to the end user.
This simplifies the set-up of data logging as a PLC-based solution, making it easy and cost effective to implement a data collection and monitoring system on a per-machine basis. This makes a decision to implement a data logging system a scalable one, rather than a costly all-or-nothing exercise, paving the way for incremental and continuous production improvements.
T: 01707 276100