The Case for Utility Boiler Fuel Delivery System Upgrades

A vital part of any coal-fired unit is its fuel delivery system (FDS). A newly formed subcommittee of the ASME Research Committee on Energy, Environment, and Waste has investigated potential FDS upgrades on three typical 500-MW wall-, tangential-, and cyclone-fired boilers. The subcommittee has produced a series of suggested upgrades that have a simple payback of no more than two years.

The American Society of Mechanical Engineers’ Research Committee on Energy, Environment, and Waste (RC EEW) was formed more than 40 years ago with a focus on industrial and municipal solid waste. The Fuel Delivery System Subcommittee was recently formed to expand the RC EEW’s original charter to include all fuels, including the energy and environmental aspects of those fuels. The first project undertaken by this subcommittee, begun in September 2011, was a feasibility and economic analysis of potential upgrades to Powder River Basin (PRB) coal-fired power plants. A summary of results of the subcommittee’s work to date follows.

Identify Plant Categories

The first step in the subcommittee’s analysis of fuel delivery systems (FDS) was to identify the family of plants of interest. A recent article (“Predicting U.S. Coal Plant Retirements,” May 2011, available in the POWER archives at powermag.com) noted that the U.S. coal-fired fleet consisted of 1,105 units with a total nameplate capacity of 342 GW at the time the article was published. A majority of those plants were between 20 and 85 years old; only 35 new plants had been added over the past 15 years.

As a group, the units 50 years and older constitute about 53 GW or 20% of the total fleet capacity and 40% of all coal-fired units—many of which may be retired due to either normal business decisions or the cost of mandated retrofits of new air quality control systems (AQCSs). The next age group, the 30- to 45-year-old units, represent 216 GW and 63% of the current coal-fired fleet. Many of these were built during the 1960s and are much more likely to invite investment in plant upgrades (Figure 1).

1. Coal fleet average unit nameplate rating. The average unit rating was calculated by averaging the rating all of the units within each age category. Data are from early 2011. Source: POWER and Burns & McDonnell

The boilers of the 30- to 45-year-old units are mainly of opposed wall-, cyclone-, and tangential-fired configuration with average capacity factors ranging from 61.8% to 73.3%, as shown in Figure 2. In this age group, there were about 226 opposed wall-fired, 143 tangential-fired, and about 15 cyclone-fired boilers in operation in the U.S. in 2011.

2. Coal fleet average capacity factor. The average unit capacity factor was calculated by averaging the reported capacity factor of all the units within each age category. Many of the units in the five years or less category did not have data available. A 75% capacity factor was estimated. In all categories, if capacity factor data was not available, that unit was omitted from the average. Data are from early 2011. Source: POWER and Burns & McDonnell

These units—the backbone of the baseload coal-fired fleet—will bear the burden of ensuring that the usual high standards of electrical grid performance, availability, and reliability are met in the future. Though most of these units have high-grade AQCSs, they will require upgrades to comply with maximum achievable control technology, but the cost is not forecast to adversely impact unit competitiveness in terms of generation cost. However, the additional AQCS upgrades required for environmental compliance will add additional complexity to plants now straining to maintain unit availability and capacity factor.

A vital part of any coal-fired unit is its fuel delivery system, as shown in Figure 3. For the purposes of the subcommittee’s analysis, the FDS consists of the feeders, pulverizers (mills), classifiers, coal piping, and burners. These systems are vital for efficient and reliable plant operations but also require substantial maintenance due to the abrasive nature of coal.

Table 1. A comparison of possible fuel delivery system upgrades and their benefits. Source: The Fuel Delivery Subcommittee of the ASME Research Committee on Energy, Environment, and Waste

Identify Possible Fuel System Upgrades

In recent years, improvements in monitoring equipment have led to significant performance improvements in FDS equipment and therefore the availability and reliability of a plant. Many of these have been in flow-measuring devices for enhanced control of fuel (such as feeder coal flow and pulverized coal flow in coal pipes) and airflow (including primary and secondary air, pulverizer preheated air, coal pipe air, wind box air, and individual burner air—both secondary and tertiary).

In a properly instrumented system, the amount of coal and air sent to individual burners so vital for low-NOx and CO operation can be measured and monitored, ensuring that good combustion takes place. Please note that cyclone boilers utilize a different FDS that consists of feeders, crushers, and cyclone burners. These differences will be addressed in the project economics section of this article.

Upgrades to the FDS to improve plant operating economics are numerous and often site- and boiler-specific (Table 1). Each FDS component upgrade can have benefits, most of which can be quantified. An example is the retrofit of a dynamic classifier, which improves coal fineness and virtually eliminates the coarse coal particles (>50 mesh). Coarse particles are a main cause of fouling and deposition in the furnace and the convection sections of the boiler. They also impede good low-NOx burner performance. Improving fineness also reduces unburned carbon in the fly ash, thus improving combustion and boiler efficiency.

Table 1. A comparison of possible fuel delivery system upgrades and their benefits. Source: The Fuel Delivery Subcommittee of the ASME Research Committee on Energy, Environment, and Waste

Additional Assumptions

Members of the subcommittee decided that three boiler types, representative of the 35- to 50-year-old fleet, would be selected: opposed wall-, tangential-, and cyclone-fired boilers. The group also decided that representative boiler systems would form the basis of the upgrade analyses, although the plant itself would remain anonymous.

A common factor shared among boilers of this vintage is that they were normally designed to burn eastern or Illinois Basin coal and have been or are being considered for conversion to a subbituminous PRB coal for emissions reduction. Most boilers have had some burner modifications over their operational life, and new modifications would be considered more for performance improvement than for emissions reduction. In focusing on the FDS, it was further assumed that air pollution retrofits would not be part of the FDS upgrades.

It was also assumed that any FDS upgrade would not increase the heat input over its original design rating and that there would be no increase in emissions, to avoid the need for a New Source Review. In addition, emissions would be less than 100 tons/year for individual pollutants, thus not triggering the Prevention of Significant Deterioration process. It was assumed that the opposed wall and cyclone boilers had selective catalytic reduction (SCR) systems installed for NOx control, but that the tangential-fired unit did not yet require installation of an SCR due to the inherently low NOx emission characteristics of that boiler design.

The analyses further assumed the plant capacity factor for this study is 80% (7,008 hours/year). In the current low natural gas price environment, this 80% capacity factor is likely higher than many units are currently operating at. A sensitivity analysis was performed using a range of lower capacity factors. The subcommittee believes that higher capacity factors will likely return as natural gas prices slowly but steadily rise in the future.

Determine Upgrades, Benefits, and Costs

For each case study, the FDS components were identified and upgrades were proposed. Next, the potential benefits of the upgrades were discussed and evaluated. The list of proposed upgrades was the source of spirited discussion among subcommittee members, although a consensus was always reached that reflected the collective expertise and diverse experience of the members.

For each case study, the upgrade’s scope was defined in sufficient detail so that reliable estimated costs could be prepared. Fortunately, the various suppliers and consulting engineers on the subcommittee were able to provide estimates of the installed costs for each upgrade considered based on prior experience.

Assessing the savings produced by the upgrades was also reached by consensus over the course of several subcommittee meetings. Difficulties arose when the economic benefit of upgrading one component alone was not possible because some upgrades were contingent on other upgrades. Also, there were interactions between the upgrades that were difficult to quantify. The savings that accrue are not always merely the sum of the individual upgrades. For example, an evaluation of the economics of using neural networks within the boiler control system produces improved boiler performance, but much of the performance improvement came from the burner modifications required by the neural net system.

Three Case Studies

The subcommittee believed that each case study 500-MW boiler must be representative of the boiler design class. Selection of the case studies was based in large measure on the spectrum of boiler types that currently constitute the U.S. utility boiler fleet, as shown in Figure 4. In considering the ~500-MW units in the 35- to 45-year age range, the vertical- and front wall–fired boiler were eliminated on the basis that few, if any, would be 500-MW-size boilers.

4. Coal fleet boiler design. The three case studies were selected from the three largest categories of boiler design now used in the fleet. Fluid bed units were not included because those fuel systems are unique and often handle opportunity fuels. Source: POWER and Burns & McDonnell

Cyclone boilers provide an interesting opportunity for evaluation in this study. Cyclone boilers lost favor in the 1980s when they were characterized as high NOx emitters and not amenable to combustion modifications. Supposedly, they were also not amenable to burning PRB coal. Today there are approximately 60 cyclone-fired boilers still in operation, of which 10 are in the 400- to 600-MW size range and four are greater than 600 MW. Many of these boilers have been successfully converted to burn PRB coal, usually with retrofitted overfire air (OFA) ports installed to aid combustion tuning for reduced NOx production.

The subcommittee decided to include a cyclone-fired unit as a separate case study. However, the FDS boundary was expanded beyond the individual cyclone feeder to beyond the coal conveyors, to the crusher-feeder island, usually located some distance from the boiler. It was felt that the crusher, more than any other device, controlled particle sizing, and the upgraded feeder provided a more uniform flow of coal to the crusher, improving overall crusher performance.

Case Study 1: The Opposed Wall–Fired Boiler

The candidate opposed wall–fired, natural circulation boiler was originally designed for eastern coal and now burns a PRB coal. It has a retrofit SCR and the original electrostatic precipitator; plans are in place to retrofit a wet scrubber. The proposed FDS components for upgrade include replacement of original feeders, retrofit dynamic classifiers on vertical shaft pulverizers, coal pipe flow-metering devices, burner modernization, retrofit OFA, airflow metering devices, and retrofit boiler control system (BCS) with a neural network. Several of the suggested upgrades are described below and summarized in Table 2.

Table 2. Case Study 1: Opposed wall–fired boiler FDS upgrades benefits and costs. The investment breakeven point was 15 months. Source: The Fuel Delivery Subcommittee of the ASME Research Committee on Energy, Environment, and Waste

Fuel Feeders. The original, roughly 40-year-old volumetric feeders were replaced with new gravimetric feeders with improved metering. They also provide some increased capacity, because PRB, with its lower heating value, requires more material throughput than the original eastern bituminous coal. Six new feeders were each estimated at $75,000 x 2 (cost of installation) = $900,000. Benefits were included with those of dynamic classifiers (DCs), below.

Pulverizers. DCs are used to increase coal fineness at a given coal throughput. In so doing, DCs also reduce the coarse grind, the 50-mesh material, which is a primary cause of slagging and fouling. A lesser-known feature of DCs is that they can also increase pulverizer capacity (depending on DC speed), fineness at higher speed, and throughput at lower speed. Typically, this is accomplished with little or no increase in pressure drop.

Because PRB is a volatile coal, there is little need for an increase in coal fineness, but there is a need for increased pulverizer throughput over the original eastern coal design conditions. DC costs were estimated as 6 x $300,000 x 2 (cost of installation) = $3,600,000, acknowledging that $300,000 is on the high side and normal installed cost is usually less than double the equipment cost.

A switch from eastern to PRB coal and its associated lower heating value means that greater quantities of PRB coal passing through the FDS and the vertical shaft mills were required to produce the same heat input. In this plant, heat input to the boiler was reduced due to coal throughput limitations. Thus, the DCs’ ability to recover lost pulverizer and boiler capacity, especially when one or more pulverizers are out of service for required maintenance work, is vital. A conservative value of 5% loss in boiler capacity was used. The reduction of 50-mesh coal particles and improved combustion was estimated to recover two days of operation at full load. Other benefits included improved load response, improved coal drying, often less vibration, but modest NOx and unburned carbon (UBC) reduction. The savings from the 5% capacity recovery was $8,760,000, and two days of full-load operation at $0.05/kWh was $960,000, for a total savings of $9,720,000.

Coal Pipes. The subcommittee considered coal pipe flowmeters because coal flow measurements may not be exact. The individual flow meters allow comparison of coal flow between pipes to ensure more equal and consistent coal flow to the burners that will in turn ensure good air to fuel ratios at each burner.

There was some discussion about replacing coal pipes on the basis of increased pressure drop or fan limitation. It was felt that coal pipes are seldom replaced, so this option was discarded. In sum, upgrades studied included the coal pipe flow modifications, $500,000, and primary airflow modifications, $200,000, for a total of $700,000.

It is vital to measure both coal and air as a precursor to aggressive efficiency improvement. It was also felt that the DCs would resolve a potential pressure drop issue, and the pressure drop issue was further clarified, as shown below in the BCS section.

Burner Modernization. The existing low-NOx burners (LNBs) are a third generation; a fourth-generation upgrade is planned that will include burner modernization and airflow-monitoring devices. Overfire air is also included. NOx reduction with these new LNBs and OFA would be about 10% (~0.02 lb/106 Btu), but they serve primarily to reduce NH3 consumption by the SCR. UBC would not drop, and CO would be held to <100 ppm.

The combination of retrofit DCs and LNBs would reduce slagging and fouling, improve flue gas flow to the SCR and air heater, provide better combustion, and improve boiler performance. Typically, LNBs are upgraded every six to eight years due to new design improvements. Costs were estimated for the LNBs as $75,000 x 24 burners x 2 (cost of installation) = $3,600,000; OFA $25,000 x 8 + $50,000 x 8 (cost of installation) = $600,000; electrical $1,000,000 (cost of installation); and $200,000 for burner management (such as new scanners and cabinets) for a total of $5,400,000. The only direct benefit was some NOx reduction that reduced annual NH3 usage by $210,000. The other benefits for improved combustion were improved boiler performance, shown in the next section.

Boiler Control System. In a detailed analysis by a Burns & McDonnell team, all the various upgrades were examined and critiqued with a review and modification of the cost when merited. In the original consideration of upgrades to the BCS it was felt a new BCS would provide benefits as well as a neural net system. As it turned out, the Burns & McDonnell team revised the BCS upgrade to only a neural network at a cost of $300,000. This applied to pulverized coal–fired boilers (Case Studies 1 and 2) and provided unexpected benefits.

Two good rules of thumb were useful in this analysis. First, “10% excess air = 0.5% boiler efficiency” and “10% excess air = 22% fan power.” Using these rules of thumb, the difference is (22.8% – 16.1%) = 6.7% change in excess air, which results in approximately 0.34% improvement in boiler efficiency and 15% improvement (decrease) in fan power.

The savings due to boiler efficiency improvements can be estimated as 500 MW x $0.05/kWh x 7,008 hours/year x 0.0034 = approx. $600,000/year of cost savings. Assuming 2 x 4,000 horsepower (hp) fans, 8,000 hp x 0.7457 kW/hp x $0.05/kWh x 7,008 hours/year x 0.15 = approx. $314,000/year savings in fan power.

Improving NOx from 0.17 lb NOx/106 Btu to 0.15 lb NOx/106 Btu will decrease the amount of ammonia consumed. Assuming 10,500 Btu/kWh heat rate, $400/ton anhydrous ammonia, SCR outlet NOx of 0.04 lb/106 Btu at 80% capacity factor, the reagent savings are approximately $85,000/year.

Project Management and Engineering Services. Engineering services, including commissioning costs, were estimated as about 15% plus an additional 10% for project management costs, which amounted to $2,725,000.

Thus, the total installed cost is estimated as $13,625,000 and the total savings are estimated as $10,925,000—a simple payback of 15 months, as shown in Table 2. It is important to note that 83% of the cost savings were the recovery of a presumed 5% unit derate taken when the unit was converted from eastern bituminous coal to a western subbituminous PRB coal.

Case Study 2: Tangential-Fired Boiler

The 500-MW tangential-fired boiler used in the analysis is a single furnace with five levels of burners and five pulverizers, 100 MW per pulverizer. Many boilers of this size and age do not have SCRs, but they often have lower NOx and UBC than other boiler designs. The subcommittee consensus was that burner modifications would be appropriate regardless of whether or not an SCR was added.

The first improvement was separating the burner levels to further reduce NOx emissions. The benefits of DCs with regard to recovery of lost capacity apply, but the need to improve slagging and fouling may not be as great. Burner modifications would involve changing buckets and adding separate overfire air (SOFA). NOx reduction may be less, but NOx emissions are likely to be low. Neural networks would provide similar improvements. Several of the suggested upgrades are described below and summarized in Table 3.

Table 3. Case Study 2: Tangential-fired boiler FDS upgrades benefits and costs. The investment breakeven point was approximately 13 months. Source: The Fuel Delivery Subcommittee of the ASME Research Committee on Energy, Environment, and Waste

Feeders. The upgrades were much the same as in Case Study 1 but with five new feeders, costing $750,000 installed. The cost benefits were included with DCs.

Pulverizers. The cost of the upgrade follows the calculations presented in Case Study 1, or $3,000,000. The calculation of benefits likewise followed the methods used in Case Study 1. Savings from the 5% capacity “recovery” were $8,760,000 and two days of full load was $960,000, for a total savings of $9,368,000.

Coal Pipes. As in Case Study 1, unit upgrades chosen were coal pipe flow and primary airflow measurement at a total of $584,000.

Burner Modernization. The modernizations comprise some bucket replacement and the addition of SOFA ports, which would require some boiler pressure part modifications. SOFA was estimated at $4,800,000. Benefits were some claimed NOx and UBC reductions, but no significant savings were estimated.

Boiler Control System. The same upgrade of a neural network with no new BCS is estimated to cost $300,000 with the similar savings forecasted in Case Study 1: boiler efficiency, $600,000, and fan reduced power costs of $314,000. There are no NH3 savings because there is no SCR. Thus, the total estimated savings are $914,000.

Project Management and Engineering Services. A consistent 25% of the upgrade costs was included, which amounts to $2,359,000.

Total cost of the upgrades is $11,793,000, with a total savings of $10,630,000. The breakeven point for the investment is approximately 13 months, as shown in Table 3. As before, 82% of the savings are attributable to recapturing 5% of the unit capacity lost as part of the initial fuel switch from an eastern bituminous coal to a lower heat content subbituminous PRB coal.

Case Study 3: Cyclone-Fired Boiler

The cyclone boiler investigated for this case study was originally designed (in the late 1960s) to fire Illinois Basin bituminous coal, but it now burns PRB coal (since the late 1980s). It has been retrofitted with OFA (1990s), SCR, dry flue gas scrubber, fabric filter, induced draft fans (converted from pressurized to balanced draft), various boiler and turbine modifications, and new O2 analyzers (2000s). Several of the suggested upgrades are described below and summarized in Table 4.

Table 4. Case Study 3: Cyclone-fired boiler FDS upgrades benefits and costs. The investment breakeven point was approximately 18 months. Source: The Fuel Delivery Subcommittee of the ASME Research Committee on Energy, Environment, and Waste

Fuel Preparation System. The feeder-crusher “island” was included in the FDS boundary, because this equipment is a vital part of coal sizing necessary for good combustion. It was generally agreed that some modification would have been made to the feeder-crusher island when switching to PRB coal, which would have included dust control and other safety-related issues. The same would have been included in all the coal conveyors, but the coal conveyors from the feeder-crusher island to the cyclone are not part of the FDS.

As part of the FDS upgrade, some upgrades will be made to the feeder-crusher island to improve coal grind at a given or increased throughput, and these costs were estimated. In discussing the cyclone modification, it was agreed that the 12 cyclones, feeders, and burners would have been previously modified, but that the upgrades to reduce UBC, which is in some cases 20% to 30%, would be required. Upgrades include a new posimetric feeder, fine grind cage crusher, and motor upgrades at $1,200,000.

Cyclone Modernization. Cyclones were upgraded with new split secondary air dampers and damper actuators. Benefits include reduced NOx and UBC by improving combustion (allowing low excess air operation); more even flue gas distribution in the furnace convection sections, SCR, and air heater; and reduced slagging. Upgrades were estimated at $2,800,000. The benefits were reduced operation at lower loads with individual cyclones forced out of service due to cyclone slag buildups and downtime for boiler deslagging. Longer time is usually required to cool cyclone furnaces for deslagging and maintenance work.

The savings estimates are based on an additional seven full days of operation in a year by reducing forced outages for cyclone cleaning. Modification costs were calculated at $3,360,000.

It was noted that if the cyclones themselves were nearing the end of their useful life and were to be replaced, there are numerous upgrades that should be incorporated in the replacement cyclones to further enhance PRB coal firing. However, if the cyclones were not to be replaced, these pressure part upgrades would not be made just to improve PRB coal firing. For this reason the costs of new cyclones or cyclone pressure part upgrades were not addressed by this study. Also not included was the use of iron oxide additives for improved slag flow.

Boiler Control System. Only an upgrade to the BCS was required—no BCS and neural network additions. This upgrade provided savings similar to those shown for Case Study 1, except there was no significant NH3 savings. The savings again related to efficiency ($298,000) and fan operation ($157,000), for a total of $455,000.

Project Management and Engineering Services. A consistent 25% of the upgrade costs was included, which is $1,120,000.

The total upgrade cost is estimated as $5,600,000, total savings as $3,815,000, and the breakeven at 18 months, as shown in Table 4. Again, note that 88% of the cost savings are attributable to potential recovery of lost generation due to derates or forced outages caused by convective pass slagging.

The Effect of Capacity Factor

The case study estimates are all based on a capacity factor of 80%, as noted earlier. While the study was under way, the increased use of natural gas reduced coal plant average capacity factor during 2012 in many regions of the U.S. Rising natural gas prices have pushed some utilities to increase coal-fired generation as the least cost option in early 2013. Table 5 illustrates the analysis results for a range of capacity factors, assuming the investment cost remains constant. Even with capacity factors in the range of 60%, the breakeven point for each project is less than two years.

Table 5. Impact of capacity factor on plant retrofit economics. Note the simple payback for each case study remains less than two years for a capacity factor down to 60%. Source: The Fuel Delivery Subcommittee of the ASME Research Committee on Energy, Environment, and Waste

The subcommittee solicits comments on the results of the analysis and suggestions for additional study. Please forward your comments and suggestions to Subcommittee Chair Robert Sommerlad.

Robert E. Sommerlad (rsommerl@aol.com) is a consultant and Subcommittee Chair. Donald B. Pearson is secretary of the PRB Coal Users’ Group. Grant E. Grothen is principal, Burns & McDonnell. Steven McCaffrey is with Greenbank Energy Solutions Inc. Other members of the Subcommittee were Robert Chase, Terrasource Global; Blaz Jurko, Gebr. Pfeiffer Inc.; David J. Stopek, Consultant, Sargent & Lundy LLC; Melanie Green, CPS Energy; Richard Himes, EPRI; Tony Licata, Licata Engineering Consulting and Chair RC EEW; and Todd Melick, Vice President, Promecon USA.