[pullquote]Pinch technology invariably gets more attention when energy or the environment are key drivers for plant design. After all, while integrating energy throughout the plant is very complex and can snarl startup, it can reduce energy and environmental footprints.
However, changes to such an integrated process can result in hard-to-understand disruptions. Blaming the last change is easy but deeper probing may be necessary to discover the real cause.
Consider the problem that afflicted a plant that was built about 40 years ago during a national energy crisis. Its design integrated as much energy recovery as possible. The plant employed four distinct heat sinks — river water, cooling-tower water, chilled water and low-temperature brine — that also were integrated. Each heat sink also could become a heat source for lower-temperature process streams.
[callToAction ]
The plant expanded several times over the years, including adding cooling tower cells and another chiller. All was good. Then the phase-out of chlorofluorocarbons happened (See the latest updates to such phase-out efforts: “Montreal Protocol Addresses HFCs.”
The chillers for water and brine were changed to an alternative refrigerant. Suddenly, problems arose. The obvious culprit was the chillers’ modification. The plant did a performance check on the chillers that led to the discovery of fouling, so the condensers were cleaned. High levels of sediment were found in the cooling water system. The basin was nearly silted in. The silt had damaged the vertical turbine pumps, so the pumps had problems supplying adequate flow to both the process and the chillers. The cooling tower fill had become silted in as well and so was replaced.
By this time, engineers were busy trying to figure out what was going wrong and how to expand capacity to meet a pickup in demand. The models seemed to indicate that heat sink capacity should have been adequate, despite the operating problems. People were puzzled. They planned projects to add capacity to the tower and to the chillers. However, such expansions required significant capital.
So, the plant formed a six-sigma team. It checked models, downloaded and analyzed data, and reviewed the process and instrumentation drawings. The team felt comfortable with its knowledge of the system. Then, interviews with the operators began.
One heat exchanger system that dictated the cooling-tower water pressure was located at the top of the tallest column on the site. That system would switch from cooling tower water to chilled water, depending on the temperature of the cooling tower water, to maintain the overheads rate. The system included separate exchangers for condensing the column overheads. A cooling-tower-water control valve adjusted the flow. By monitoring the pressure upstream of the valve, operators knew when to either add or shut down a pump to maintain the pressure required.
The review with the operators began with the process control engineer talking about how the system controlled the overheads pressure. As the discussion proceeded, one of the operators interjected: “That isn’t how it works!” He then explained how the controls really worked and how operators contended with some problems that this caused. A team was sent out to check how the valve and controls operated. Sure enough, the valve — still the original one installed 40 years ago — acted exactly opposite to the way specified.
The signal to close actually was opening the valve, raising flow until it was limited by available pressure. This meant the chilled-water cooling stream was increased more to offset the effects of heating on the process condensing stream —starving flow to other parts of the process as well as the condensers on the chillers; the excess water flow back to the cooling tower effectively limited the ability of the tower to reduce the temperature.
Simply reversing the valve action provided more than enough capacity in the cooling sinks to allow a major increase in capacity, obviating the projects to boost capacity of the cooling sinks and millions of dollars in capital expenditures.
The key takeaway here is to check your assumptions when diagnosing problems with complex systems, especially those that are highly integrated like most pinch-technology designs. (For details on potential control problems with such designs, see: “Control Challenges Can Pinch Energy Savings.”)