Legacy Updates for Plants & Processes

One of the greatest hardware-related problems connected to an economic downturn is the decline of infrastructure. Our current recession has been an example of this, particularly in process plants, and the effects of the decline will persist long after business has begun picking up. Companies that are working on thin margins in the best of times find that a downturn forces them to reduce costs e...

By Peter Welander, Control Engineering December 1, 2009
For more information, visit:
Hear more from Mustang at www.controleng.com/podcasts

One of the greatest hardware-related problems connected to an economic downturn is the decline of infrastructure. Our current recession has been an example of this, particularly in process plants, and the effects of the decline will persist long after business has begun picking up. Companies that are working on thin margins in the best of times find that a downturn forces them to reduce costs even deeper, so operators and maintenance personnel are cut, and plant improvements are put on hold. Although older systems require more maintenance, maintenance tends to get deferred and breakdowns soon begin to interfere with production. Declining production then prompts more cutbacks.

Unless a company makes a conscious decision to invest for the long term, it’s a downward spiral until the process can no longer run or the plant is deliberately closed when economic viability ceases. This unfortunate scenario has happened all too often in recent years. One of the better possible outcomes is that a distressed plant might be improved as part of a larger project to revamp or repurpose it to manufacture a different product slate. But if a company has excess capacity, it might be sold off or kept in mothballs and then called back into service if circumstances change.

There are many possible scenarios that could put you in a situation of having to resurrect a plant or process unit that’s been operating at minimal effectiveness or shut down altogether. Reversing a declining spiral in your own company or inheriting someone else’s mess through acquisition can be a major challenge.

Investing in a legacy plant is a major business decision within any context. In all likelihood, your company management has already decided what it will require from a given plant or unit to make the project practical. By the time you get involved, the commitment has already been made, so your task is finding a way to make the plan a reality. Of course, the operators and maintenance people who knew how things really worked may have been lost to cutbacks, so where do you start if you’re coming in cold?

Field survey

The first step is to gather up all the information and data you can on the plant equipment, configuration, and its past performance. Obviously the more current the better, but even old information has some value. With that in hand, the next step is walking through the plant and seeing what’s there.

David Martindale, director of refining, engineering, and services for UOP, has been involved in many such projects in the oil industry. “Let’s say a customer is looking to purchase an entire refinery. We look at the units’ key factors and the performance to see where the gaps are,” he says. “Typically the owners or potential owners have done that to some degree. They’ll do a bit of benchmarking on their own, and then they’ll come to us and say, ‘I want to run this type of crude, and I want to make these products with these specs. What’s missing?’ Usually they have some ideas, and we come up with additional ideas on revamping key units or even adding a unit.”

But is performance and equipment information always available? Plants that have to operate shorthanded aren’t always the best when it comes to keeping information current. Those that are in perpetual fire-fighting mode may not update records when a system has been patched back together to run another day.

Martindale has faced that very situation. “We can do two things,” he says. “First, we can usually get the major equipment information. There will be some specification for a heater or a vessel, a compressor or whatever. Sometimes it’s many years old, and doesn’t necessarily reflect the current condition of the equipment, but it tells us something about the size and rating of that piece. The other is to have our staff go on site to help develop the process flow diagram, the P&ID, as necessary to understand the process flow and what equipment is there. A lot of things are covered up by insulation, so it isn’t always easy. But are there scenarios where there are no P&IDs, no product flow diagrams, no equipment information or specifications? In my experience, we’ve never run into a situation that bad.”

Evaluating automation

Along with the process equipment, the automation systems and hardware have to be evaluated as well. This includes the main control system, field instrumentation and actuators, wiring, and other infrastructure. An evaluation has to include the equipment’s current condition, as well as its potential to handle whatever product the management may want to produce after the upgrade or repurposing.

Looking at the automation infrastructure involves two phases. The first begins at the enterprise level and works its way down. The second deals with the details of field devices, wiring, and the nuts and bolts that make the plant run.

Top down

Unless you work for a very small company within a plant that is essentially isolated, there will be demands from management for information as to what is going on in the upgraded operation. This means the automation system will have to be integrated into enterprise level systems to get data where it needs to go. This is nothing out of the ordinary, since this kind of data moves around in all sorts of companies; but when coming into a new operation, such procedures may not be in place and need to be implemented.

One of the first steps in this process is finding out who needs what information. This can involve extensive discussions with a company’s management. Don Colchin, president of Mustang Automation and Control, compares the process of analyzing these relationships and lines of responsibility to creating a P&ID for the process. In the same way that feedstocks and additives have to go through reactors, information has to be pulled out of the system at strategic points and sent to the right individuals. As he puts it, “We bring in technology that can improve how that client does business and gets information into different parts of the organization. We’re starting at that point to marry the functional world with the physical world of hardware and software.”

This analysis phase is also a good time to consider where improvements can be made to work processes and systems to improve efficiencies and cut costs.

Jorge Motes, Honeywell Process Solutions director of program management for the Americas, works with company managers at the beginning of an upgrade project in a format he characterizes as a workshop to identify critical system needs at the outset. Once the study is underway, there are many opportunities to identify possible improvements and build these into the program.

“From those discovery discussions, we start evaluating based on what they want to get out of it, the budget and other constraints, and what kind of solution can we provide,” says Motes. “We once looked at a customer that was still completely relay based, and there were a lot of manual reports being used. We talked to them about this and demonstrated how their investment could be recovered in less than a year, as their costs change positively when shifting from manual to automatic, and how that information can help them.”

In the trenches

While enterprise-level discussions are important, you can’t get data from a DCS (distributed control system) that won’t run. At some point you have to get in the trenches to check individual instrumentation transmitters, marshalling cabinets, and I/O cards. That can be an eye-opening experience depending on how old the equipment is and how far it has deteriorated.

First, contact the supplier of the installed control system, no matter how old it is, suggests Ken Keiser, PCS7 marketing manager for Siemens. “We and other vendors can provide audits,” he suggests. “In many cases they’re either inexpensive or free depending on the situation. We can look at what they have and, at least from the control system side, say this isn’t going to work or this is too far gone, but this over here you can keep. We help you fill in the missing parts of the control system.”

Obsolescence is a major factor, particularly with older pneumatic devices and early electronic equipment. Moreover, earlier systems were not usually protected as well once distributed control architectures allowed monolithic systems to be broken up and placed around the plant. Many early installations did not use sophisticated enclosures to keep out dust and corrosive fumes. Circuit boards also lacked coatings leaving traces exposed. Older disk drives may not have been protected adequately against dust.

Once you’ve determined what will and won’t work, there needs to be some judgment calls as to how much of the older equipment you want to bring back into operation. Keiser advises, “You can make product with a lot of old equipment, but whether it’s secure enough, reliable enough, and the data can get through to third-party applications well enough, that’s another question. You have other needs beyond just getting product out the door. There are internal IT security needs and other enterprise data needs that an older system can’t fulfill. What’s the IT department going to say when you’re trying to resurrect a Microsoft Windows 2000 machine? Sometimes you just have to recognize that it’s too old a system.”

Challenging cost cutting

If you’re in a situation where you are trying to reverse a declining spiral spurred by calls for cost cutting, one strategy may be to make the case for improving profitability by increasing production and efficiency rather than cutting costs. This will almost certainly require investment which may not be easy to sell to management, so you will have to prepare a thorough study with all the specific costs and production projections you can compile.

The corporate culture within your company may suggest how such a request will be received. Some companies appreciate such entrepreneurial efforts more than others. In any case, your ability to explain how you will attack the challenge will go a long way to support your requests.

Successfully turning around a plant and making it into a productive and profitable operation through effective use of automation and people resources runs counter to so much of what’s actually happening these days. Anyone who wants to lead such a charge will be putting his or her career on the line, but pulling it off could be hugely rewarding.


Author Information
Peter Welander is process industries editor. Reach him at PWelander@cfemedia.com .