Why are there missing measurements?

CONTROL magazine logo

by Paul Studebaker

Today’s refineries--and process plants in general--are missing the measurements they need to bring millions of dollars to the bottom line through improved reliability, higher energy efficiency, lower labor costs and risk management.

 “These missing measurements are not on the process flow or P&I diagrams,” said Gary Hawkins, senior refining consultant for Emerson Process Management and former 31-year veteran at UOP, in his Thursday morning session at Emerson Exchange. “Most process units are designed and built with the minimal amount of instrumentation--only that which is required for safe operations, to validate process design.” Manual acquisition is used for long-term trends because measurements add to project cost, and adopting new technology takes time.

“At UOP, we had the EPCs design and build to spec, and the unit only had to meet the specs until the end of validation. We got a seven-day warranty--if after the first week the unit meets the design requirements, the start-up crew leaves.” So the measurements needed for validation are taken by gauges and temporary instrumentation, not permanent transmitters.

Typically missing are the measurements you need to monitor the condition of the equipment--temperature, pressure, flow and vibration readings you can use to improve site safety, prevent outages and product losses, and reduce maintenance costs of equipment such as pumps, heat exchangers, cooling towers, steam traps and relief valves.

On typical pumps backed up by spares, “When you suggest it might make sense to monitor temperatures, pressures and vibration, you get told, ‘That’s why there are spares.’ Wouldn’t you prefer to know before the pump catches fire, before the BLEVE--the boiling liquid expanding vapor explosion?”

On heat exchangers, with temperature and flow measurements, you can automatically calculate and monitor duty and fouling factors, and know when to schedule a cleaning. This is increasingly important with today’s crude oils. “Shale oil is often highly paraffinic and unstable when blended with conventional crudes, which is causing intermittent and rapid increases in fouling rates,” Hawkins said. You can compensate for lost exchanger efficiency up to a point and at higher energy cost by firing up the preheaters, but “You can only get so much heat, then you have to reduce capacity.” A costly situation when a refinery is selling all it can produce.

On rotating equipment, “The $20 million dollar machines are protected, but lesser compressors are also critical to operations. Process availability justifies the cost.” In one example, just saving 30% of a 0.2% production loss due to failures can save a refinery more than $300,000 per year. “And that 30% could easily be 80%,” Hawkins said. “These are very conservative numbers.” Vibration switches give too many false alarms and not enough information, Hawkins said. "It’s better to use vibration transmitters."

Air-cooled exchangers can use a combination of rotating equipment (vibration and temperature) and heat exchanger (temperature and flow) monitoring to see problems early and save repair and downtime costs--there Hawkins' example saves $207,000 per year.

Even cooling water towers can offer significant paybacks. For vibration monitoring, “Handhelds work well but how often do they get up to those fans?” Hawkins asked. “The environment is wet, slippery, sometimes icy. It’s treacherous to get those readings.” His cooling tower example included costs of energy, water, water treatment chemicals and disposal, and saves $459,000 per year--without maintenance or productivity considerations.

Some missing measurements weren’t even available when plants were built, such as acoustic monitoring of steam traps and relief valves. Here you not only can save on steam (when traps fail open), you also can ensure process integrity when traps fail closed, and minimize releases, product losses and safety incidents by knowing immediately when relief valves open or leak.

But nobody is going to add the “missing measurements” for you. “It needs to be end user-driven,” Hawkins said. “It doesn’t benefit the designer or process licensor or the site contractor. It’s up to you, the user, to see the benefits and demand the measurements.”

Hawkins suggested that both old and new plants install wireless networks to minimize the cost and provide flexibility in adding new measurements. “Prioritize areas and pieces of equipment where they will do the most good,” he said. Many plants look first at the tank farm, which can be a huge investment. Look at your plant on Google Maps--the process area typically is not that big, and it doesn’t take much network equipment to cover it.

Hawkins concluded, “Ten years from now, we’ll look back and most refiners will have installed this second layer of automation.”

  • In my personal opinion, this presentation really embodies what Peter Zornio called “pervasive sensing” at the opening of the Exchange. It was a very good session.

    WirelessHART enables existing sites to be modernized & sustained in ways previously not possible, benefitting both operations and maintenance. Wireless changes how we do things, for the better; how we maintain assets, how we look at energy consumption, how personnel operate in the field, and how we manage unmanned sites. One day every plant & field will have 1,000 wireless transmitters. Some sites are already very close.

    EPCs can guide clients to modernize the plants they built years ago as a new business model to tie them through slow periods between projects. There is a formal process to modernize a plant step by step using wireless sensors. New projects should also deploy plant-wide wireless infrastructure, to make sure the new plant is not built to be run & maintained the old way with operator rounds; don’t build a new plant old.

    With additional sensors throughout the plant, software, and operating procedures in place, "desktop maintenance" planning, energy conservation, HS&E compliance, and integrated operations can become a reality. Assets fitted with additional sensors become smart assets able diagnose their conditions and communicate it to operations & maintenance personnel. With these additional sensors in place, operator rounds can be reduced since the data comes to the operator, the operator need not go to the data. In other words, automate not only control loops, but also automate operator rounds and maintenance inspections. The process equipment is inspecting itself; a new type of self-diagnostics. “Pervasive sensing” across the plant is fundamental to this goal.

    Adding sensors using wireless instead of wires enables accelerated deployment making it possible and low risk to deploy during a short turnaround or even while the plant is running. Another element of this is the non-intrusive mechanical installation possible for many of these measurements; strap-on (ultrasonic sensor), bolt-on (position sensor), clamp-on (temperature sensor), stick-on, screw-on, or magnetic (vibration sensor) eliminate the need for additional process penetrations for these measurements. This dramatically reduces cost and risk of deployment as there is no need to cut, drill, or weld pipes and vessels. Moreover, many of the measurement do not need high accuracy; they just need “good enough”.

    Therefore more and more plants are installing WirelessHART coverage throughout their process unit areas. Plant modernization is not just system migration, but to add automation where there was none before. Indeed, it is not necessary to replace the control system to add wireless sensors. Most plants add wireless sensors to their existing system and many of the new wireless sensors may connect to separate system for reliability, non-process safety, environmental, energy, and security – separate from the process control and safety system.

    Many operation and maintenance problems around the plant can be solved by deploying WirelessHART transmitters “beyond the P&ID” together with an asset management system and other software to obtain asset health information and other plant data. Wireless enables plants to innovate and experiment with new measurements to solve old operational problems.

    Today we look back 20 years and wonder how we managed to get things done without a mobile phone, let alone a smart phone. 20 years from now we will look back and wonder how we could run a plant without wireless transmitters.

    Learn more about pervasive sensors for missing measurements beyond the P&ID using a second layer of automation from this article: