Page 5 - Industrial Technology EXTRA 21st June 2021
P. 5

Originally, most plant operators took a reactive   first and most obvious is that we want to check the
       approach, where essentially a device was allowed to   device’s health. If it is going wrong, perhaps
       break and then was fixed. This was an expensive   suffering a fault that is causing it to stray outside its
       method as in certain cases it could lead to the   calibration parameters, this could affect production
       whole plant tripping out, extended down time and a   quality by causing problems such as inaccurate
       rush to find replacement parts or devices. For   blending, dosing or billing. The other reason, even
       example, in a batch production environment, such   more important, is ensuring that processes remain
       as mixing orange juice concentrate with liquid in a   in a safe condition.
       food and beverage plant, the failure of a flow meter
       would not allow the right proportions to be mixed to  Using the data
       produce the correct quality of product and the   At a very basic level, taking advantage og these
       process would need to stop until the problem was   capabilities comes down to how we communicate
       resolved.                            with the device. Although most measurement
         This is bad enough in the above batch method   devices are digital, many applications continue to
       but in a petrochemical process, based on 24/7   use the tried and tested 4-20mA current loop rather
       continuous production, allowing a temperature   than digital communications networks. For
       transmitter to simply fail could be potentially   example, all ABB’s 4-20mA measurement devices
       catastrophic. Things got better in the 1970s to 80s,   also offer HART communications, a bi-directional
       when preventive maintenance came to the fore,   communication protocol that provides data access
       with devices maintained to a schedule. Even this   between intelligent field instruments and host
       was highly manual, with equipment being checked   systems. The communications link carries the
       by staff walking the line to note which devices were   process values, like flow rates or temperature.
       reading or behaving as expected and which ones   It also gives access to the device’s diagnostics
       needed attention.                    data, such as its operational status. Many advanced
         Over the last few years, the concept of predictive   digitally enabled measurement devices will offer
       maintenance has become ever more popular, with   sensor checks including internal connections, an
       the aim of trying to prevent failure events in the   indication of sensor memory failure and an internal
       future. Although predictive maintenance is still a   electronics check. With communications in place,
       form of preventive maintenance, it uses actual   device interrogation is the next step. There are two
       condition-based data to determine when to perform   methods for interrogating device data.
       maintenance, as opposed to a pre-determined   The first is inbuilt condition monitoring and
       schedule based on historical information.   diagnostics based on the NAMUR NE 107
         By using condition monitoring of equipment, we   standard. To make things easier for technicians and
       can know more about exactly when to schedule   other process personnel who need to deal with
       maintenance and prepare for it in advance, for   alarms, NAMUR NE 107 categorizes internal
       example, ordering spare parts. We can also make   diagnostics into four standard status signals —
       use of the increasingly rich data available from   failure, function check, out of specification and
       measurement devices to identify exactly what went   maintenance required, each of which can also
       wrong and derive strategies or techniques to prevent   contain greater detail. For example, with a failure
       the same problems occurring in the future. There   signal, can the failure be traced to the device, or
       are two main reasons for condition monitoring. The   alternatively, is the process itself at fault?

                            June 2021 • INDUSTRIAL TECHNOLOGY EXTRA • p5
   1   2   3   4   5   6   7   8   9   10