Coal, Emissions

Combining Ultrafiltration and Reverse Osmosis Technologies Key to Producing Pure Water for Power Plant Boilers

Issue 1 and Volume 116.

By Brian Wise, global product manager, industrial systems and Robert Banks, global product manager, GE Power & Water

As power plant designs have evolved over the past 50 years, top boiler pressure i.e. superheated high pressure turbine inlet conditions, has increased dramatically. As pressure levels increased, the required boiler feed water quality increased proportionately. Today’s power stations routinely require extremely high quality water (as defined by EPRI and their European counterparts) to protect the turbine and steam generator. Impurities to be controlled typically include all measurable cations, anions, silica, organics that may exist in source water as well as dissolved gases.

In the last several decades, reverse osmosis (RO) plus mixed bed ion exchange or electrodeionization (EDI) has become the primary technologies used in place of—or as a pretreatment step to—conventional cation bed plus anion bed plus mixed bed ion exchange systems due to the competitive capital cost, lower operating cost, and decreased health and safety concerns with ion exchange (IX) regeneration chemicals. The concern that most end users have regarding RO technology is that these systems are prone to fouling and not as robust as chemically regenerated IX systems, especially when the water is from a surface water source. Advances in membrane filtration technology (ultrafiltration or microfiltration) for RO pretreatment have improved the efficiency, quality of feed water and reliability of downstream RO systems. These improvements are resulting in more displacements of and pretreatment additions to IX systems with RO technology to dramatically reduce operating costs, reduce the water treatment system’s footprint, improve safety and extend equipment life.

Operating Overview with Ion Exchange Demineralizers

IX demineralizers typically are configured as a Twin Bed Ion Exchange Demineralizer (TBDI) followed by a Mixed Bed Ion Exchange Demineralizer (MBDI). These IX systems have been in place for decades as the primary technology to purify raw water to meet the requirements of high quality / high pressure makeup.

A typical IX system can have cation, anion, and mixed bed exchangers containing a mixture of both anion and cation exchange materials. Cation exchangers work by removing positively charged ions from water known as cations and exchanging them with hydrogen ions. Anion exchanges work by removing negatively charged ions from water known as anions and exchanging them with hydroxide ions. The hydrogen and hydroxide ions combine to form pure water. Operating this way for some time, the ion exchange resin is exhausted with cations and anions from the water stream and they need to be “regenerated.” The cation resin is regenerated with a strong acid (typically sulfuric acid) and the anion resin is regenerated with a strong base (typically sodium hydroxide). After the regeneration process the resin is returned to a state where it can again remove cations and anions from the water source and produce purified water. The regeneration chemicals are hazardous and require large volumes to be stored on site.

Reverse Osmosis Gains Acceptance as Viable Alternative to Traditional IX

In the last 20 years, RO has gained acceptance as an alternative to this traditional IX method as concerns over safety of the regeneration chemicals, the waste that is generated from the regeneration process and the high operating costs become more prevalent. Today, it is estimated that 70 percent of newly installed water treatment systems in power plants are using RO technology-plus-MBDI rather than TBDI-plus-MBDI.

Operating Overview with RO Plus IX

Reducing or eliminating the frequency of regenerating IX materials has led to the increase in popularity of non-IX based water filtration methods such as RO systems, which use membrane technology to facilitate the purification of water. Pressure is applied to the water entering an RO system to force some of the water to pass through a semi-permeable membrane resulting in two streams of water leaving the RO system. The water passing through the RO membrane as product or permeate water has approximately 99 percent of dissolved ions removed, leaving behind a second stream of water with more concentrated level of ions as the waste stream. The ratio of permeate water produced is typically 75 to 80 percent of the feed water volume, leaving 20-25 percent of concentrated wastewater. RO machines are typically used ahead of IX systems to reduce the ionic load on the IX system.

An important consideration when contemplating RO technology—in addition to or in replacement of existing IX systems—is that the feed water to the RO systems requires substantial pretreatment to remove suspended materials and organic materials. The mineral concentration of the RO feed water is also important to ensure the anti-scalant treatment scheme can control mineral precipitation in the RO concentrate stream at high water recovery ratios. The typical feed water quality guidelines for an RO system are shown below:

Turbidity: < 1NTU
Silt Density Index (SDI): < 3 (15 minute test)
TOC: < 5 mg/L
Total Suspended Solids (TSS): < 1 mg/L
Iron: < 0.05 mg/L

For well water supplies, these values are relatively easy to achieve with conventional multi-media filtration. However, for surface water sources, these values can be difficult to consistently achieve with a conventional clarification system using coagulant and flocculant chemicals with gravity sand filtration and multi-media filtration. The results for the RO system can be frequent chemical cleanings and reduced RO membrane life, making the operating expenses higher than desired.

To address such concerns, ultrafiltration (UF) membranes have gained in popularity in the last decade as a pretreatment step to RO because they can deliver consistently better quality water to the RO system as compared to conventional multi-media filtration. UF membranes such as the GE ZeeWeed 1500 are backwashable filters with a 0.04 nominal micron rating and 0.1 micron absolute as compared to a multimedia filter with a nominal 20 micron rating. The typical UF product water quality will be consistently at 0.2 NTU turbidity or less, 2 SDI or less, and have 50-90 percent of TOC removal which will meet the RO feed water quality guidelines. TOC removal is a special bonus of UF systems. Without UF surface water, TOC may foul anion resin or pass all the way into the mixed bed, and possibly into the steam cycle. Once in the steam cycle, TOC has been known to decompose to low molecular weight organics, thus making it difficult to control boiler water buffer conditions.

Advantages of Deploying PROPAK System to Offer UF Pretreatment to RO Membranes

One such system that takes advantage of UF and RO membrane technologies is called the PROPAK. This machine as developed by GE Power & Water utilizes UF as pretreatment to RO membranes on one skid in a relatively compact footprint. The system’s filtration skid and tank skid provide a 35 percent footprint reduction compared to typical conventional pretreatment and IX systems producing similar flow rates. The system is excellent for taking surface water sources and achieving consistent results of equal quality compared to a traditional TBDI system with significantly reduced chemical consumption. The PROPAK is also an excellent addition as a pretreatment step to existing IX treatment systems to drastically reduce the consumption of regeneration chemicals and lowering the overall operating expense of the boiler feed water treatment system.

FirstEnergy PROPAK Installation Showcases Significant Cost Savings for Operators

Utilizing membrane water filtration followed by RO will save in chemical consumption, increase plant safety, and improve water quality consistency. An existing IX system at FirstEnergy, Hatfield’s power plant in Pennsylvania was requiring regeneration every other day from raw water feed with conductivity ranging between 300 and 450 microsiemens per square inch (uS) conductivity. The plant wanted to add RO technology followed by mobile mixed bed polishing to replace the IX system, reduce operating costs and remove the bulk acid and caustic storage, but the source water is from the river with high seasonal fluctuations in turbidity and TDS. Adding UF technology as pretreatment was desired to help deal with the feed water fluctuations but there were capital and floor space constraints on the project. In 2010 FirstEnergy selected a GE PROPAK unit as the utility’s solution to pretreat the mobile mixed-bed system.

The PROPAK unit selected was configured as a UF plus two-pass RO to deliver 130 gallons-per-minute of permeate, and Propak is unique in that all technologies are mounted on a common frame. The compact PROPAK unit fit in the available space and adding a PROPAK system delivered two-pass RO permeate with less than 2 uS conductivity to the mobile mixed beds. As a result, at an average cost of $1,333 per regeneration, the plant will save approximately $200,000 annually in chemical costs alone. With the elimination of on-site regeneration cycles, the station was able to remove sulfuric acid and caustic regeneration tanks which reduced chemical handling and also reduced sulfate and TDS loading to the wastewater treatment system.

The PROPAK is designed for variable surface water sources to give the customer a reliable system that is easy to install while requiring a relatively small footprint. The UF plus two pass RO configuration consistently provides high quality water reducing operating expenses. The patented integrated three purpose tank skid and automation package gives PROPAK the performance advantage to help end users maximize the savings in their plant.

Safely Operating Current Transformers

By Alan P. Sappe’, Product Marketing Lead Specialist, Industrial Connection Technology

It can take a lot to bring down a major electric system. It required a magnitude 9.0 earthquake followed by a 46-foot wave, for example, to bring down the 40-year-old Fukushima Daiichi nuclear power plant belonging to Tokyo Electric Power Co. in Japan. And if disaster-related transportation problems hadn’t delayed an executive, emergency steps might have been taken in time to prevent the meltdown. Of course, a more modern facility design would have been able to shut down safely. Even so, out of the more than fifteen thousand deaths in the country, none occurred at Japan’s 55 reactors.

While the United States hasn’t witnessed a serious nuclear incident for many decades, the country has experienced failures such as the August 2003 blackout that affected 50 million people in the U.S. and Canada. That event was largely down to the cascading effect of a series of faults and errors. Smart operators, therefore, are paying close attention to the maintenance of each of the components in a power plant or distribution system. Current transformers (CTs), for example, if treated right, will provide reliable, trouble-free service. Treated wrong, however, and CT’s can generate a deadly fireball.

CT Basics

Tools do not exist to directly measure and monitor high levels of current. Current transformers are used to generate a secondary low-amperage current, without affecting the main current flow, so that ammeters, wattmeters, VArmeters and relays can be attached.

The general design for a current transformer is similar to other transformers with a primary winding, secondary winding and a magnetic core. What distinguishes current transformers is that, rather than providing power to a load at a specified voltage or amperage, their main function is for the secondary circuit to accurately duplicate the status of the primary circuit, but at a far lower current. In the U.S., it is standard for the secondary circuit to be at 5 amps, and in other countries at 1 amp. There are also CTs with milliamp outputs for use with microprocessor-based meters and relays. To achieve the desired result, the most common design, relatively unchanged for the last 80 years, is for the secondary circuit to be wrapping wire around a silicon-coated steel ring hundreds of times. The primary circuit is a single turn – the main power conductor passing through the center of the ring. The alternating current in the primary circuit induces a magnetic field in the core, which produces a much smaller current in the secondary circuit. With this design, to achieve maximum accuracy, the primary circuit should be directly in the middle of the secondary circuit.

CTs are used throughout the electrical system from the power plant to the end user, wherever measurement, control and safety devices are used. For low voltage applications, such as measuring the flow on a circuit servicing a piece of equipment such as a pump, blower, or conveyor, a CT will just be installed over the power conduit. For high voltage distribution lines leaving the generator or at substations, the CT will need to be properly insulated from the ground.

They are rated according to the amount they step down the current (a 2000:1 CT would produce a 1 amp flow for every 2000 amps in the primary circuit), the maximum voltage rating, voltage range, insulation class and accuracy. Some CTs are built with several taps at different ratios. Those that are used to bill customers need higher levels of accuracy than ones that are part of a protective circuit. Standards for CTs include IEC 60044-1, Edition 1.2 (2003) Instrument transformers –Part 1: Current transformers which “applies to newly manufactured current transformers for use withelectrical measuring instruments and electrical protective devices at frequencies from 15 Hz to100 Hz” and IEEE/ANSI c57.13-2008 IEEE Standard Requirements for Instrument Transformers. IEEE also has an unapproved draft standard covering primary circuits above 115kV.

Each phase three phase circuit would have its own CT and a single primary circuit may have more than one attached. For example, one metering-class CT may be used to determine power usage for billing or to provide data to a plant control system, while a second would be used to trigger overload protection systems.

CT Operation and Maintenance

CTs are there to provide a safe way to measure circuits, but only if proper safety procedures are followed regarding maintenance and operation of the CTs. There are two aspects that must be watched – the transformers themselves and the terminal blocks. CTs are subject to the same failure mechanisms as other transformers including short circuits, dielectric breakdown and bushing failures and so should follow the usual inspection and maintenance routines as executed for other types of transformers.

The Electric Power Research Institute’s (EPRI) Power Transformer Maintenance and Application Guide discusses issues connected with both potential transformers (PTs) and current transformers used for instrumentation and says that PTs have a longer service life compared to CTs, and when CTs fail, they fail catastrophically, and may cause losses and damage. For voltages less than 22,000 V, the transformers use a dry or compound filling and those over 22,000 V use a compound or oil and the heat generated by the current causes the oil to degrade. Poor quality oil, poor quality paper insulation, oil contamination, excess temperature and damage from switching transients were all found to cause failure of instrument transformers.

The other safety aspect deals with ensuring the secondary circuit is always closed. As the EPRI transformer manual states:

“Because the secondary circuit must be set, the secondary should never be open circuited because without the closed circuit path, the CT would act as a step up transformer with a very high magnetizing current with an exceptional amount of flux in the magnetic core. Therefore, a CT should always have its secondary shorted when it is not connected to an external load in order to avoid dangerous and destructive arcing.”

If the secondary circuit is opened during service, or the primary circuit is brought on line without the secondary circuit being closed first, the arcing can produce a violent fireball. Even with lower voltage and amperage CTs such as are used for metering equipment inside a plant, an open secondary circuit can damage the CT and any instruments connected to it, as well as injure plant personnel.

Building Better Blocks

In addition to maintaining the transformers, the terminal block is another potential source of catastrophic error. The secondary circuit passes through the terminal block and connects to the instrument leads at that point. There are times, however, when the instrumentation circuit must be broken, for example to replace the instrument or to send it a calibration signal. In these cases, it is essential that the secondary circuit be shorted at the terminal block before disconnecting the instrument.

The typical procedure is to insert a screw to short the transformer at the block, disconnect the measuring circuit from the secondary circuit using a knife disconnect or other method and insert the test probes or signal generator. The steps are reversed when completed.

While the current at the terminal block is relatively low – five amps for the U.S. and one amp elsewhere in the world – the danger lies in omitting a step, doing it in the wrong sequence, or failing to connect the right leads to the right terminals on the block, resulting in an open secondary circuit.

To minimize the chance of error, and make it easier on the technicians, advances have been made in the design of CT shorting terminal blocks. Instead of inserting a screw to short the circuit, the blocks come with disconnect levers which clearly show by their position whether the circuit is open or closed.

To reduce the chance of wiring errors, the blocks are also designed for clear, professional labeling showing exactly what wire should be connected at each position. This makes it easy to make the correct connection, rather than having to count the number of positions before threading the wire into the block.

For convenience, blocks are now available which allow ring lug connections to be made using either a screw or a bolt, whichever matches the type of installation tool the technicians prefer.

From a design standpoint, the shorting blocks can now be installed on DIN rails, instead of having to be mounted to the back of the instrument panel. This gives greater flexibility for putting the connections where they are most convenient without space constraints.

Even the best design, however, won’t do the job safely unless personnel are trained to follow safety guidelines. But the right terminal blocks make it easier for them to do the job and minimize the odds of human error.

About the Author: Alan P. Sappe’ is Product Marketing Lead Specialist for Industrial Connection Technology at Phoenix Contact USA, A provider of industrial connection technology, automation technology, electronic Interface systems and surge protection.

More Power Engineering Issue Articles
Power Engineerng Issue Archives
View Power Generation Articles on