Coal, Gas

Repowering Revisited

Issue 11 and Volume 119.

By S. C. Gülen

In 1995, the Manchester Street repowering project in Providence, Rhode Island, was the largest repowering project at the time. The project boosted the station’s output from 132 MW to 489 MW. What’s more, the station’s heat rate improved 25 percent. Siemens supplied the gas turbines and ABB supplied the steam turbine. Photo courtesy: Jef Nickerson


S.C. Gülen is Principal Engineer at Bechtel Infrastructure Power

According to a recent Power Engineering news article, a total of 60 GW (60,000 MW or approximately the equivalent of 120 advanced F or H class (50 Hz) 1×1 gas turbine combined cycle power plants) of coal-fired generation is expected to retire by 2020, a number beyond that previously reported to the U.S. Energy Information Administration (EIA).

At the end of 2012, there were 1,308 coal-fired generating units in the U.S. totaling 310 GW of capacity. In 2012 alone, there were ~10 GW of coal capacity retired (~ 3 percent of the 2011 total). Units that retired in 2010-2012 were small, with an average size of 97 MW, compared to those retiring over the next 10 years, which are an average of 145 MW.

According to EIA’s Annual Energy Outlook 2014, 90 percent of those retirements would happen by 2016 due to the implementation of the Mercury and Air Toxics Standards (MATS), which took effect in April 2015. According to the 2015 report, compliance with MATS, low natural gas prices and competition from renewables are projected to lead to the retirement of 31 GW of coal-fired generating capacity and the conversion of 4 GW of coal-fired generating capacity to natural gas between 2014 and 2016. Note that these projections do not include the impact of EPA’s proposed Clean Power Plan (e.g., Section 111(d)) or other actions beyond the existing policies. This plan, if implemented, would increase the numbers cited above drastically.

In a report published by Union of Concerned Scientists (UCS) [2], using economic criteria, a range of 153 to 353 coal-fired electric utility generating units (from a national total of 1,169) were identified as ripe for retirement. These coal units are good candidates for closure because they are economically uncompetitive compared with cleaner, more affordable energy sources. They collectively represent 16.4 to 59.0 GW of generation capacity and 1.7 to 6.3 percent of total U.S. electricity used in 2009 (the most recent year of available data at the time of the report’s writing).

The potential closure of these units would be in addition to the 288 units representing 41.2 GW of coal-fired generating capacity already scheduled by their owners for closure, which produced nearly 4 percent of U.S. electricity use in 2009. Together, the ripe-for-retirement units plus the already announced closures would constitute a combined 100 GW of potential coal plant retirements.

Like the announced retirements, the coal generators that are ripe for retirement are typically older, less utilized, and dirtier than the rest of the nation’s coal fleet. The owners of these soon-to-be-retired generators have concluded that paying for costly upgrades to keep their outdated coal plants running is a bad investment-particularly now that there are many cleaner, lower-cost alternatives that can replace old coal units while maintaining the reliability of the electric system. Whether natural gas, clean renewable energy from the wind and sun, or cost-effective efficiency measures to reduce electricity use, energy options that are abundant, cheaper, and cleaner are making it harder for coal to compete.


While the UCS report claims that every region of the country has the potential to replace the generation from the retired generators, such as increasing the use of renewable energy and ramping up underused natural gas plants, it does not mention the most obvious one, repowering, as an option.

Repowering involves the addition to or replacement of aging power generation equipment with obsolete technology (i.e., a fossil-fired boiler plant with a steam turbine generator) with newer equipment comprising up-to-date technology (i.e., natural gas fired gas turbine in simple or combined cycle) while retaining still-usable components towards more effective utilization of an existing site, which involves one or more of the following:

  • Improve generation economics (higher output and efficiency)
  • Extend plant life
  • Improve environmental performance
  • Enhance operability and maintainability

The hurdle for repowering is the life cycle cost of generation: that is, the new power plant must have better LCOE (capital expenditure, fuel, O&M) than the old one.

This is a tall order. The aging steam power plants (the average age of ripe-for-retirement coal power plants is nearly 50 years!) are fully paid-off generating assets, whose generating costs are limited to the variable expenses for fuel (typically coal) and O&M. Unless the equipment is in such a terrible shape that O&M costs to maintain a reasonable efficiency (in order to keep the fuel costs in check) become intolerable, these plants could be operational for quite a long time (albeit at a reduced capacity factor).

The hurdle is significantly lower nowadays primarily due to much more stringent emissions control requirements enforced by government agencies that make the aging fossil power plants impossible to keep running. The investment into new equipment to reduce harmful stack emissions and improve the efficiency is simply too high. The situation is made even worse by low natural gas prices driven by increased production of shale gas using new technology such as fracking. Thus, natural gas fired gas turbine technology with efficiencies pushing 60 percent becomes economically a much more attractive alternative.

Types of Repowering

There are four major gas turbine-based repowering options:

  • Site repowering (a completely new gas turbine combined cycle (GTCC) power plant)
  • Heat recovery repowering, i.e., boiler replacement with a gas turbine plus HRSG (Heat Recovery Steam Generator) train
  • Hot wind-box repowering (utilizing a GT as “forced draft fan”)
  • Feed-water heater repowering (utilizing GT exhaust gas for feed-water heating)

Only two of the four possible types of GT-based repowering options are applicable to repowering of retired coal fired power plants:

  1. Site replacement (a new brown-field GTCC project)
  2. Replacement of original boiler with a GT+HRSG train (heat recovery repowering)

Heat recovery repowering is where the existing boiler is replaced by a GT and a HRSG. This approach increases the unit’s net generating capacity by about 150-200 percent, reduces the heat rate by up to 30-40 percent and reduces NOx emissions. The performance change is application specific and depends on the match of the new gas turbine with the existing power plant. Since heat recovery repowering delivers the biggest “bang for the buck”, most of the existing repowering experience in the U.S. involves this system type. Due to the relatively large capacity increase, this approach is normally considered for older units less than 250 MW with steam pressures up to 1,800 psia.

This type of repowering requires deciding between optimizing the existing steam turbine performance with the new combined cycle components and installing a new steam turbine. A closely related decision associated with the first choice is whether to retain the existing feed-water heaters or not (i.e., maximize steam turbine output). Optimizing the older plant performance is important for the repowered unit to be able to compete with a new unit, even if it has lower capital costs. New steam turbines, main transformers and other equipment add to the capital cost, but may be justified by gains in output, efficiency or to provide reliable operation.

There are myriad factors that enter the decision-making process in repowering of an existing coal-fired steam turbine power plant. An incomplete list is given below:

  • Capacity factor
  • nstalled capacity per fuel in the service area
  • Cost of fuel
  • Net plant efficiency (heat rate)
  • Cycling requirements (integration with renewable generation)
  • Size of the power plant
  • Other operating costs
  • Anticipated life
  • Outstanding or anticipated air quality control system (AQCS) equipment upgrades
  • Location of the power plant
  • Community support for or against the plant
  • Other

Similarly, there are myriad factors involved when considering natural gas-fired gas turbine based repowering options. (Note that one way to implement feed-water heater repowering is to utilize solar thermal energy instead of a gas turbine.) Some of those factors are listed below:

  • Age of the steam turbine and anticipated life (for GT+HRSG replacing the boiler)
  • Steam conditions (efficiency) of the steam turbine
  • Size/capacity of the steam turbine
  • Power demand vs. installed capacity
  • Cycling requirements
  • Other operating costs
  • Other

A recent cost estimate for different repowering options from 2013 is shown in Table 1.

Which Repowering?

The heat recovery option is extremely appealing in terms of reduced engineering complexity and construction risk, minimum life cycle cost as measured by LCOE and maintaining base load generation capability with minimal additional capital investment. The assumptions leading to this assertion are as follows:

  1. The steam turbine of the retired coal-fired plant is in good shape (age, performance, etc.)
  2. The heat rejection system (condenser and cooling tower) is in good shape
  3. The switchyard is in good shape and can be expanded to accommodate the second generator’s output

The definition of the term “good shape” above is as follows:

The equipment in question can be continued to be used as-is (1) with minimum repair and/or upgrades and (2) without major redesign and/or retrofitting to a capacity above or below their original rating.

While the first item in the definition above is obvious, the second item needs some elaboration. What is meant by that clause in terms of specific equipment is this:

  1. The steam turbine will operate in a regenerative cycle with the same steam conditions (e.g., 1,450 psia throttle pressure and 1,000°F for main and hot reheat steam) and same steam flows with all feed water heaters operating as designed. In other words, there will be no plugging of existing bleed holes and/or drilling of new ones for additional steam admission. The unit will generate the same output at its rated conditions.
  2. The condenser will operate with the same cooling water flow and temperature rise to condense the same amount of steam as designed at its rated conditions.

The assertion of minimal additional capital investment to maintain the base load generation capability can be justified by simple arithmetic. Let us assume the following

TRCC = Total coal-fired generation capacity to be retired

NGUC = New generation unit capacity to replace TRCC

RPF = Fraction of TRCC amenable to the proposed repowering scheme

Thus, we have

NGUC = TRCC – (1 + a) x RPF x TRCC

NGUC = TRCC ´ [1 – (1 + a) x RPF]

where a is the increased generation capacity multiplier applied to the steam turbine capacity of the repowered coal-fired plant to account for the gas turbine. In other words, if the coal-fired boiler of a fossil plant with 100 MW net rated output is replaced by an F class GT (say, General Electric Frame 7) and mildly-fired HRSG, the net rated output of the new (repowered) combined cycle power plant becomes (1 + a) x 100 MW. The expected (average) value of a is 200 percent or 2.0 (see the example below).

Consequently, with no repowering at all, i.e., RPF = 0, NGUC = TRCC. If only 10 percent of TRCC is amenable to repowering, i.e., RPF = 0.1, NGUC = 0.7·TRCC.

In other words, if TRCC estimate is taken at its high value of 60 GW, only 10 percent repowering saves 0.3×60 GW = 18 GW = 18,000 MW of NGUC; 6 GW from keeping the steam turbine generators of the retired plants and 12 GW from the repowering gas turbines.

Assuming all of that NGUC is new GTCC at $1,000/kW, the replacement cost will be

Cost (new GTCC) = 18,000×1,000x$1,000/kW = $18 billion

Assuming that the extra 2.0×0.1×60 GW = 12 GW = 12,000 MW (from GT+HRSG+BOP for repowering) comes at $750/kW, the cost will be

Cost (Repowered) = 12,000×1,000x$750/kW = $9 billion

Thus, the savings in upfront capital investment is $9 billion!

An Example Calculation

Let us assume that the coal-fired power plant slated for retirement is a subcritical pulverized coal (PC) unit rated at 120 MWe. Although a fictitious plant, in terms of size and performance, it is representative of ripe-for-retirement plants.

The reheat steam turbine (similar to General Electric’s A series fossil steam turbines with separate HP casing and combined IP/LP casing) has five (5) feed water heaters and operates at relatively modest steam conditions (i.e., 1,450 psia / 1,000°F / 1,000°F). The closed-loop heat rejection system comprises a water-cooled condenser (about 1.8 inches of mercury pressure at ISO base load) and wet cooling tower. There is no IP or LP steam addition. The “new and clean” gross (turbine) heat rate of this unit is 8,436 Btu/kWh (~ 40.5 percent) with 127 MW generator output.

According to a recent article, available plant data identified several steam turbine units nationwide rated between 100 and 200 MWe and installed between the mid-1960s and mid 1970s [5]. One such station was equipped with two units rated at 110 MWe with gas fired boilers and reheat steam turbines (208.3 lb/s throttle steam flow and 1,900 psig/1,000°F cycle) at a gross turbine heat rate of 9,800 Btu/kWh (about 35 percent efficiency).

Thus, applying 10 percent heat rate degradation might result in a more representative output (about 116 MW) and heat rate for the sample unit herein than the new and clean value. There are two ways to alleviate this shortcoming:

  1. The HRSG and the duct burner (DB) are sized to achieve 127 MW steam turbine generator (STG) output with the degraded ST power island.
  2. Steam path of the existing STG is upgraded to a new and clean condition.

In either case, lost performance is recovered at the expense of additional capital investment.

The goal is to utilize the STG along with its regenerative feed-water heat exchanger train (including the boiler feed pumps) and the plant’s heat rejection system with no changes, i.e.

  • Same steam cycle conditions
  • Same HP steam flow
  • Same extractions for feed-water heating
  • Same condenser back-pressure
  • Same STG output at rated conditions

The fossil boiler is decommissioned and replaced by a single-pressure HRSG, which utilizes the exhaust gas of an advanced F class GT (e.g., GE’s 7FA.05) and a duct burner in front of the HP evaporator section for steam generation, superheating and reheating.

The final, repowered plant arrangement is shown below. Note that the HRSG contains an economizer section for water heating. The hot water (450°F) at 650 psia is utilized to heat the GT fuel gas (assumed to be 100 percent methane) to 410°F for increased efficiency (performance fuel gas heating).

The placement of the DB upstream of the evaporator aims to avoid the added cost and complexity of oversized attemperation (desuperheating) equipment (piping, valves, spray nozzles, etc.). It is estimated to be a relatively small, 3-runner burner. (There are actually HRSGs equipped with large duct burners located upstream of the evaporator.) There could be a potential for creating significant hot spots on the evaporator surface, which in turn may cause problems with circulation. The possibility is higher in a smaller duct burner. Nevertheless, if approached carefully, such a duct burner can be properly designed.

The DB fuel gas flow (same as the GT fuel) is controlled to achieve the desired HP steam generation at various site ambient conditions. Another benefit of the DB placement and minimal firing (less than 1,100°F at the DB exit) is to limit the size and cost of the SCR for NOx and CO abatement. (The SCR is not included in the calculations below. It can add around $2 million to the equipment cost.)

The performance of the original steam turbine plant and the GTCC plant after repowering (at ISO conditions) is summarized in the table below. Note that performance and cost calculations are done using Thermoflow’s flexible heat balance simulation software, Thermoflex, and its companion tool, PEACE (Plant Engineering And Cost Estimation).

The marginal (incremental) efficiency, 63-65 percent, is the ratio of incremental plant net output to incremental plant heat consumption. In other words, the repowered plant generates the additional 210-220 MW of electricity (net) at a higher efficiency than an advanced H class GTCC.

The key question is the cost of the benefits provided by the repowered GTCC power plant. A rough cost estimate is given below. The equipment cost (total $80 million) is about $355 per additional kilowatt. About $2.5 million of that is either for a larger HRSG or STG steam path refurbishment. The total installed cost (TIC) is estimated at about $575 per additional kilowatt (about $126 million) excluding the demolition and removal of the existing boiler.

To put the cost and performance of the proposed repowering in perspective, consider a new GTCC power plant. The closest candidate on a total performance basis is a 1×1 plant with General Electric’s 7FA.05 gas turbine:

  • 323.0 MW net output (ISO base load – 112.0 MW from the STG)
  • 58.2 percent net plant efficiency
  • Budget plant price $224.8 million ($696 per kilowatt)

Note that the GTW Handbook performance numbers for the new GTCC plants are ISO base load values with 1.2 inches of mercury backpressure and once-through (open loop) water-cooled condenser. The auxiliary power consumption for GTW ratings is about 1.6 percent of the gross output whereas the auxiliary power consumption of the repowered GTCC of the example case herein is 2.8 percent (see Table 2). The difference is mainly driven by the heat rejection system’s parasitic power consumption.

It is quite difficult to establish an “apples-to-apples” cost comparison basis with information from two different sources and different scope. Thermoflow’s PEACE software came up with $219 million for “Contractor’s Price” of a 1×1 GTCC with GE’s 7FA.05 gas turbine, which is equivalent to GTW Handbook’s price. Adding 30 percent to $120.5 million equipment cost of PEACE, a total owner’s cost of $255 million is arrived at. As a rough first cut, this is deemed to be a reasonable comparative total cost number vis-à-vis about $126 million for the repowered plant.

Thus, the cost-performance comparison between the proposed repowered coal-fired plant and a new GTCC is summarized here:

A Sustainable Solution

The heat balance calculations unequivocally show that the performance boost is similar to what is expected from a full-blown heat recovery repowering (i.e., three-pressure reheat HRSG, IP and LP steam admission with steam turbine – and possibly heat rejection system – modifications and upgrades:

  • 200 percent increase in net outpu
  • 30 percent reduction in net heat rate

In terms of cost and environmental impact,

  • About $130 million less upfront capital investment (vis-à-vis an advanced F class GTCC)
  • $375 per kilowatt vis-à-vis $800/kW for a new GTCC. (As a comparison, consider that the FP&L’s Fort Myers project with 960 MW incremental output in 2002 came with a bill of nearly $800 per kilowatt in today’s dollars – probably more.)
  • Comparable LCOE (small advantage shown in Table 4, about 2 percent on average, is not significant enough)
  • 150,000 to 180,000 tons less CO2 emissions (70 percent capacity factor) – 21-25 percent reduction
  • Elimination of all toxic stack emissions such as mercury, arsenic, acid gases etc. (not obvious from the data herein but a fact given the difference between the two fuels, i.e., coal and natural gas, and combustion systems)

The feasibility of the proposed concept obviously hinges around the reduced capital investment and project technology risk. The latter is a known Achilles’ heel for repowering projects. Thus, the first and foremost requirement is a diligent evaluation of the equipment (age, degradation, requisite repairs, modifications and/or upgrades to ensure continued service with no excessive O&M expenditures) in the fossil-fired plant slated for retirement (and thus a candidate for repowering). It should be done on a case-by-case basis to ensure that the plant in question is suitable to the implementation of the proposed simple heat recovery repowering concept.

A bona fide economic evaluation of repowering should be based on a power generation system planning study. Such a study requires detailed simulation of how the power system generating units operate to meet the load demand over a period of time (typically 20 years).

Capacity addition decisions are made to meet the required reserve margin or a generation system reliability target. If a repowering decision is implemented, the future generation addition schedule is impacted and may result in future capacity needs. Capacity savings along with the resulting fuel and O&M savings from more efficient operation comprise the repowering benefits and savings.