                                       


                                  MEMORANDUM



Tetra Tech, Inc.
10306 Eaton Place, Suite 340
Fairfax, VA 22030
phone	703-385-6000
fax	703-385-6007

TO:			Paul Shriner and Lisa Biddle, EPA
FROM:		John Sunda, Tetra Tech
DATE: 		August 27, 2008 (revised May 13, 2014)

SUBJECT:		Variable Speed Cooling Water Pump Technology Application as a Flow Reduction Technology for 316(b) and Costs

Introduction

At most existing power plants the intake cooling water pumps are fixed speed and are sized to handle the maximum flow required. These pumps may even be oversized, as many industrial system designers allow a contingency on the system head required. (Gambica undated) Fixed speed pumps must constantly operate at the maximum flow rate regardless of the cooling water flow requirements, while the actual flow requirement will vary with the power generation rate and surface water temperature. The use of variable speed pumps (VSPs) instead of fixed speed pumps at cooling water intakes allows plant operators the flexibility to adjust the cooling water flow rate to better match the system's cooling requirements.

A reduced flow volume will result in reduced O&M costs as a result of the reduction in pump energy requirements. Depending on site-specific conditions this reduction can, in many cases, result in recovery of the capital cost of the technology over a period of time, producing savings thereafter. In fact, VSPs are often employed in industrial systems solely for their economic benefit. In the case of power plant intakes, the reduction in flow volume has the added benefit of reducing impingement and entrainment (I&E) impacts.

VSPs can be used to reduce flow volume even during periods of peak power generation, but there are operational limitations and consequences associated with this flow reduction technology. These limitations include:

             *       Inherent limits of the technology that, based on system characteristics, may restrict pump operation to a specified flow range to prevent damage to the pump. The system hydraulic characteristics will also affect the amount of savings in pump energy cost; 
             *       Limits in flow reduction associated with National Pollution Discharge Elimination System (NPDES) Permit heat limits, since a decrease in flow will result in an increase in condenser outlet temperature;
             *       Economic consequences of reduced plant generation output resulting from reduced turbine efficiency associated with higher condenser temperatures.
 
 The latter two limitations are more of a concern during periods when the source water is warmer, and will also tend to limit flow reduction during periods when the system is operating at peak capacity.

I&E Reduction from Flow Reduction 

There is no question that the flow volume reduction associated with use of VSPs can result in direct and proportional reduction in the impacts for both impingement and entrainment (I&E). For the purposes of compliance and benefits assessment, EPA has assumed that there is a direct one-to-one relationship between flow reduction and the reduction in I&E impacts. In other words, a flow reduction of 20% is assumed to result in a reduction in both I&E of approximately 20%. This may actually be a conservative assumption (i.e., an underestimate of total reduction in impact), as the reduction in the approach velocity for an intake may allow an even greater proportion of actively swimming fish to avoid the intake and escape impingement and entrainment. Also, the reduction in through-screen velocity may reduce injury to impinged organisms that are returned via fish return, by reducing the force holding them against the screen.

Flow Reduction Using Existing Fixed Speed Pumps

The typical configuration for intake water pumps involves use of one to two or more fixed speed pumps in parallel that are used to pump water though each steam condenser. There may also be a spare pump that does not operate during normal conditions. Fixed speed pumps will produce a constant volume of flow for any given set of hydraulic conditions, with flow volume being affected mostly by the pumping head. The pumping head is a measure of the energy delivered to the water by the pump and is often expressed in "feet." The pumping head is the sum of the height the water is raised plus the friction losses as water is forced through the system.

For fixed speed pumps, there are only two ways to reduce the flow volume: one is to turn off one or more pumps if there are several operating in parallel; and the other is to throttle the flow downstream of the pump, resulting in an increase in the pumping head. Both approaches have disadvantages and limitations. Turning off a pump can only be done when two or more pumps are operating in parallel and when the system conditions allow for a substantial flow reduction, like when the generators are operating well below capacity and/or when source water temperatures are very low. Throttling the flow has the disadvantage of increasing pumping head, which increases the pump energy requirements; in addition, the amount of throttling is limited by the hydraulic capabilities and characteristics of the system. These limitations substantially restrict the opportunity for flow reduction for most fixed speed pump systems.

VSP Retrofit

A VSP retrofit involves replacing fixed speed intake pumps with variable speed pumps. At a minimum, this involves the installation of a variable frequency drive (VFD) and replacement of the pump motor, switches, and controller. In many cases, this may be all that is needed. A variable frequency drive is an electronic device that varies the pump motor speed by varying the electrical frequency of the AC power delivered to the pump motor. In some cases, the existing motor may not be designed to handle the added harmonic electric currents associated with this type of system. In such cases, the pump motor may need to be derated (i.e., the maximum power output and flow rate is reduced) or the motor will need to be replaced or rewound. Additionally, the pump itself may require replacement or modification if the existing pump hydraulic characteristics are not compatible with variable speed operation and/or result in undesirable limitations on the amount of flow reduction that can be obtained. If multiple pumps are operated simultaneously and in parallel, it is best to retrofit all of the pumps.

The use of VFDs allows the flow through the pumps to be controlled over a range of flow volumes, thus allowing the flow volume to be tailored to the plant operating conditions. With proper control, the effect on turbine efficiency can be minimized and the effluent temperature can be maintained within the NPDES permit temperature limits. This allows the facility full flexibility to effect both small and moderate flow volume reductions when conditions allow.

During the winter months, use of flow reduction can actually result in an increase in turbine efficiency by eliminating subcooling in the condensers. Subcooling occurs when the steam condensate in the condenser is cooled excessively, resulting in the system's consumption of additional heat to bring the condensate back up to the boiling temperature when it is recycled back to the boilers. Excessive subcooling can also result in the formation of condensed water droplets within the last stage of the turbine, which can damage the turbine blades. Measures to control excessive subcooling include the flow reduction methods described above for fixed speed pumps, as well as piping configurations that can bypass a portion of the flow around the condensers and piping configurations that can recirculate condenser outflow back to the pump inlet. In the latter case, some flow reduction may already be occurring but without the benefit of reduced pumping energy requirements. The control of subcooling, especially slight to moderate subcooling that might otherwise be tolerated, provides another economic benefit for VSP retrofits through increased plant power output.

Using Variable Speed Pumps in Parallel with Single Speed Pumps

A potential costs saving measure would be to install a variable speed drive for only one pump of several pumps operating in parallel (i.e., discharge to a common pipe or header). 
Conceivably, this could cut costs by one-half or more while still providing considerable flow reduction flexibility. However, there are limitations that prevent this from being a practical option in many circumstances. Centrifugal pumps have operating curves where, at any given speed, the flow volume varies with pumping head. Flow decreases as the pump head increases. A common rule for pumps is that all pumps operating in parallel should run to a similar characteristic which usually means identical pumps at identical speed (e.g., two VSPs can do this as long as the speeds are similar). If a fixed speed pump is operated in parallel with a variable speed of the same size, then the two curves will become different as the speed of one pumps is reduced. A point can be reached where the two curves do not intersect and flow in the slower speed pump will stop because it cannot produce sufficient head. This places limits on the operational flow range and the interaction of the pumps will create complexities in operation and control that are best avoided. It would be acceptable for a dedicated spare pump to remain fixed speed since all pumps could be operated at full speed under circumstances when the spare is needed. In developing cost estimates, it would be appropriate to assume that all pumps would be retrofitted with variable speed drives.

Operational Limitations

There are technical limitations to the amount of volume reduction that can be achieved with VSPs. For any pump, as the speed is reduced, there is a point reached where the pump's output head is equal to the system's static head, resulting in zero flow. Continuous operation at such a condition must be avoided because the impeller will continue to spin and the water will recirculate within the pump casing, resulting in damage to the pump. The flow volume response to varying speed is unique for every combination of pump and system hydraulics, and thus the minimum safe speed must be calculated for each application to avoid operation at or even near the shutoff head. System controls are set such that the minimum pump speed will be well above that which produces zero flow conditions. Two power plants in California (Pittsburg and Contra Costa) have installed VSPs and the minimum flow that can be attained at these plants by reducing pump speed is 50% of maximum flow. (Mirant Delta 2007)

Another limitation is related to the desire to maintain a minimum average flow velocity in the condenser tubes. If the tube velocity drops below about 3 feet per second (fps), the pressure drop across the tubes becomes so low that the distribution of flow between tubes becomes less uniform reducing the predictability of condenser performance and increasing the potential for fouling. (Heat Exchange Institute 2006) Maximum recommended design tube velocities vary based on tube material, ranging from 6 fps for Admiralty steel to 8-10 fps for copper/nickel alloy and +30 fps for stainless steel. (Janikowski 2003)

Another limitation is that at certain speeds pumps may suffer from excessive vibration when the frequency of the vibration caused by rotation matches the natural resonant frequency of the pump or of the support structure. While pumps may be safely transitioned through these speeds, operation at these critical speeds for a prolonged period may cause serious damage to the pumps. VFD control systems can be programmed to transition through and avoid operating at selected ranges of pump speed.

One important system characteristic that affects the performance of VSPs is whether the total pumping head is predominantly the result of losses from friction or from static head. Where the pumping head is predominantly from friction losses, the flow reduction capability of VSPs is greater and overall system efficiency at reduced flows will be greater. An example of a system where friction losses are a large component of the pumping head would be a system that uses an inverted siphon configuration. Inverted siphon configurations are often used in once-through systems where the condenser elevation is close to the water surface, because they are well worth the savings in pump energy requirements associated with the siphon configuration. Such systems require vacuum pumps to remove the gases that collect in the high points. The height of the inverted siphon is limited to prevent the formation of water vapor under the vacuum conditions that may form within the siphon. If the condenser elevation is above the maximum siphon height, then the siphon height is shortened by exposing the downstream end to the air at an elevation above that of the source water in a structure called a seal pit. Facilities where the condensers are located well above the water surface will have higher static components of the pumping head even when inverted siphons are used. Thus, the condenser elevation and piping configuration will affect the performance of VSPs.

In systems where the pumping head is predominantly static head, as the pump speed is reduced a point is soon reached where small changes in speed can result in large changes in flow rate, especially as the pumping head approaches the system static head as described above. Thus, the available range of flow reduction is much lower than in systems where the pumping head is mostly friction losses. Also, in systems where the pumping head is predominantly static head, the pump efficiency drops substantially with reduced speed. Such systems will experience much less savings in power usage. Thus, use of VSPs in such systems is less advantageous. In high static head systems, the pump and system hydraulic characteristics must be carefully evaluated before deciding whether the benefits are worthwhile.

When the turbine system is operating at a given generation rate (i.e., a constant steam load), a reduction of the cooling water flow volume will result in a proportional increase in the condenser temperatures. This will result in an increase in the difference in cooling water temperature between the condenser inlet and the condenser outlet (delta T). Many plants have NPDES permit conditions that set a maximum limit for the delta T value. This effectively places a practical limit on the amount of flow reduction that can be achieved. During warmer months, the increase in condenser temperature will also result in a higher turbine exhaust pressure, resulting in a reduction in turbine efficiency. Thus, there is a competing economic incentive to maintain higher flow levels.

Many plants have NPDES permit conditions that set a maximum effluent temperature, which may put additional limitations on the availability of flow reduction through variable speed pumping, especially during summer months, regardless of the economic considerations. In fact, under extreme summer conditions, some plants may be required to maintain the cooling water flow at full capacity while having to reduce power output (derate) in order to meet temperature limits.

As described above, the amount of flow reduction that can be achieved has both operational and seasonal limitations. In general, opportunities for flow reduction are greater during cooler months and thus the benefits of I&E reductions may be enhanced or reduced depending on the timing of the seasonal variations in the presence and behavior of the various life stages of the affected aquatic organisms.

Applicability 

Unless existing intake velocities are relatively close to the 0.5 fps standard, flow reduction through the use of VSPs alone may not be sufficient to result in compliance with 316(b) I&E reduction standards. The technology, however, should be effective in producing partial reductions and when coupled with other technologies, may help ensure compliance. Because of the economic benefit associated with reduced pumping energy requirements, VSPs may be useful even when the other technologies are fully capable of meeting the I&E requirements by themselves and when the presence of sensitive organisms coincides with the period when the source water is cooler. VSPs should be considered as a technology option that can provide moderate I&E reductions that complement other technologies that may otherwise be unable or only marginally able to meet the I&E reduction requirements.

The capital costs of a VSP retrofit will be dependent on which components of the pumps need to be replaced; it should be assumed, at a minimum, that a retrofit will include replacement of the pump motors. Given the savings in pump energy costs associated with VSPs, the net operating costs should be negative in most applications (i.e., savings in pump energy costs will exceed any maintenance costs). Actual savings will be highly variable depending on the system hydraulic conditions, the plant operating schedule, and the degree of flow reduction attained. Under very favorable conditions, the net operating savings may offset capital costs (i.e., the technology will pay for itself). However, if flow volume reduction is aggressively sought, then pump energy savings will be offset by reduced plant output associated with a reduction in turbine efficiency.

VSPs will be most effective when:

 Facility capacity utilization rates are not very high.
 Cooling pump head is predominantly from friction losses and not static head.
 They are combined with other I&E reduction technologies.
 Existing screen velocities can be reduced to below 0.5 fps.

The amount of flow reduction that can be achieved at any given moment will vary from 0% to as high as 50% or more. Given this variability, the overall flow reduction will lie somewhere in between and will be dependent on the factors described above. Technologies that could benefit from being paired with VSPs may include:

 Traveling screens - Modules 1, 2, 2a, 3
 Fish barrier net - Module 5
 Velocity cap - Module 8

Since reduced flow volume will result in a reduction in the approach and through-screen velocities, VSPs will likely result in improved performance of velocity caps and traveling screens, particularly those with high approach velocities. If existing screen velocities are already close to 0.5 fps, then compliance with impingement requirements may be possible using VSPs alone. For example, an intake with a screen velocity of 0.7 fps could meet the 0.5 limit with a consistent flow reduction of 30%.

Costs

Capital Costs

A VSP retrofit can involve installation/replacement of the following:

   1. The variable frequency drive and associated instrumentation and controls
   2. The pump motor
   3. The pump and impeller
   4. Power source modifications
   5. Containment structure and HVAC modifications

The simplest retrofits will involve only item 1 but some projects could involve all five.
Since the power requirement is the primary cost driver for variable frequency drives, the costs are often expressed in terms of dollars per horsepower (Hp). However, since data concerning cooling water pump motor horsepower is generally not available to be used for cost estimation, costs will be estimated using dollars per gallon per minute (gpm) cooling water flow. Since the variable frequency drives will be working with the existing pumps, the design intake flow should serve as the basis. Table 1 below presents a summary of available variable speed pump total capital cost data for both installed projects and project estimates. One of the completed projects, the Pacheco Pumping Plant, is a water supply pump station with a very high pumping head of 241 feet (ft) and, thus, the dollar/gpm value does not compare well with power plants that operate at pumping heads closer to 50 to 60 ft. In order to provide comparable unit costs, the equivalent flow value for pumps with the same total motor power requirement with a pumping head of 60 ft instead of 241 ft was calculated and used to derive $11.6/gpm unit shown. All costs are adjusted for inflation to 2009 dollars. In general, variable frequency drive costs in dollars/Hp have declined over time and therefore some of the older projects (Pittsburg, Contra Costa, Brayton Point) may be overestimated. The median value of $15/gpm should serve as a reasonable estimate for a typical VSP retrofit. 

                                   Table 1.
          Summary of Capital Costs for Variable Speed Pump Retrofits
Source
                                 Capital Cost
                               Unit Cost - Flow
                                   Unit Cost
                                    - Power
 
                                 2009 Dollars
                                     $/gpm
                                     $/Hp
Brayton Point Demonstration
                                  $1,250,000
                                     $14.0
                                     $971
Port Jefferson DT&C Estimate
                                  $5,337,000
                                     $26.2
                                    $1,451
Pittsburg CA Plant 1985 Estimate 
                                  $5,749,000
                                     $14.6
                                       
Contra Costa Units 1-5
                                  $8,210,106
                                     $21.6
                                       
Pacheco Pumping Plant (241 ft head)
                                  $14,000,000
                                     $46.5
                                     $583
Pacheco Pumping Plant Adj. to 60 ft head[a]
                                  $14,000,000
                                     $11.6
                                     $583
Carlsbad Desalination
                                  $8,500,000
                                     $15.4
                                       
Median[b]
                                       
                                      $15
                                     $971
      [a] See footnote 2
      [b]Excludes Pacheco Pumping Plant at 241 ft head

O&M Costs 

Operating costs of VSPs will consist of pump energy and equipment maintenance. Maintenance for the pump and motor should not change and, in fact, may decrease due to less mechanical stress on the system at lower speeds. Maintenance costs for the variable frequency drive (VFD) are low and only a slight amount of additional energy is required to run the VFD. While some additional periodic labor by a skilled electrician may be needed, the system will likely be controlled by personnel in the main control room. Thus, the only significant change in O&M costs will be the reduction in pump energy requirements (i.e., negative net O&M). Assuming the goal is to maximize flow reduction, it is reasonable to assume that an aggressive flow reduction may be sought and, at times, will result in reduced turbine efficiency, especially during the summer. The simplest O&M cost approach would be to assume that the system will be operated such that the annual reduction in pumping energy will be roughly equal to the combined cost of variable frequency drive maintenance plus the energy penalty associated with reduced turbine efficiency (see discussion below). In other words, there would be no net gain or loss in annual O&M costs or total net power generation.

For facilities that do not require aggressive flow reduction to meet I&E reduction requirements and where site conditions are moderately favorable, this technology should be viewed as a low cost technology with respect to capital and net O&M costs. As such, it is an available low-cost technology option that could be used to improve and supplement the performance of other onsite I&E reduction technologies.

Energy Penalty

For any given steam load from the steam turbine, a reduction in cooling water flow will result in an increase in the temperature rise between the condenser inlet and outlet and a corresponding rise in the condensing temperature. This in turn increases the backpressure on the steam turbines resulting in a reduction of turbine energy output. Table 7-5 of the EPRI Technical Report "Closed-Cycle Cooling System Retrofit Study Capital and Performance Cost Estimates" presents turbine performance data for 13 different steam turbines of various designs ranging from lower to high pressure steam. The average calculated heat rate penalty from this table expressed as a percent of plant output is shown in Table 2. (EPRI 2011) The relationship between turbine backpressure (water vapor pressure) and condensing temperature is shown in Figure 1 along with an equation fitted to the data that allows the calculation of vapor pressure for any given water temperature. The heat rate penalty for any given increase in condensing temperature will increase with increasing initial temperature (i.e., marginal heat rate is higher at higher water temperatures). The increase in the slope of the line in Figure 1 as the water temperature rises is part of the reason for this. Table 2 also presents a rough estimate of the incremental rate of heat rate penalty increase for each increase in turbine backpressure. For example, as the condensing temperature increases from 93 degrees Fahrenheit ([o]F) to 102 [o]F the average increase in the heat rate penalty will be 0.03% per [o]F.

                                   Table 2.
                     Summary of Average Heat Rate Penalty 
                             Turbine Back Pressure
                            Condensing Temperature
                               Heat Rate Penalty
                          Inches of mercury (in. Hg)
                                     [o]F
                           % per In. Hg Increase[a]
                            % per [o]F Increase[b]
                         Total Penalty From 1.5 In Hg
                                      1.5
                                      93
                                     0.6%
                                     0.03%
                                      0%
                                       2
                                      102
                                     1.5%
                                     0.09%
                                     0.3%
                                       3
                                      114
                                     2.1%
                                     0.2%
                                     1.8%
                                       4
                                      123
                                     2.3%
                                     0.3%
                                     3.9%
                                       5
                                      130
                                       -
                                       -
                                     6.3%
[a] Average incremental rate per 1.0 In. Hg if temperature/back pressure is increased from this point.
[b] Average incremental rate per 1.0 [o]F if temperature/back pressure is increased from this point.

Assuming that the condenser heat rejection rate (flow x temperature rise) remains constant, a flow reduction of 20% for a cooling system with a temperature rise of 15 [o]F would result in a temperature rise of 18.75 [o]F or an increase in the condenser outlet temperature of 3.75 [o]F. As a result, the condensing temperature inside the condenser will have a similar rise. At an initial condensing temperatures of 93 [o]F, the corresponding energy penalty would be approximately 0.11%. At an initial condensing temperature of 102 [o]F, the corresponding energy penalty would be approximately 0.34%. The corresponding values for a larger flow reduction of 30% would be 0.2% and 0.6%, for initial condensing temperatures of 93 and 102 [o]F, respectively. The condensing temperature is the sum of the source water temperature, the condenser temperature increase (range), and the condenser total temperature difference (TTD) which is defined as the difference between the temperature of the hot water and the steam condensate exiting the condenser. The TTD typically ranges between 3-4 degrees Celsius ([o]C) (5.4 to 7.2 [o]F) but can be higher for condensers designed to operate with a consistently cold water source. For a condenser with a TTD of 6 [o]F and a range of 15 [o]F, the cooling source water temperature corresponding to condensing temperatures of 93 [o]F and 102 [o]F would be 72 [o]F and 81[o]F, respectively.

A typical cooling flow requirement for a once-through fossil-fuel cooling system with a range of 15 [o]F would be approximately 600 gpm/megawatt (MW). Assuming a pumping head of 60 ft and a pump efficiency of 75%, the corresponding pumping energy requirement will be about 1% of plant output. If flow is reduced by 20%, the corresponding energy savings would be 0.2% of plant output. This pumping energy savings is similar in value to the energy penalty when the source water temperature is around 80 [o]F in this example. During periods when the source water is cooler, for VSP installations with relatively low flow reduction values, in the 0 to 20% range, the pumping energy savings should be equal to or greater than any energy penalty. For higher flow reductions, especially during periods when source water is warmer,the energy penalty may need to be factored into economic analyses.

                                   Figure 1.
             Relation of Condensing Temperature to Steam Pressure

Note: The equation calculates the vapor pressure for any given temperature in the range shown. In the equation y = water temperature in degrees F (Deg F) and x = vapor pressure in In Hg. R2 is the coefficient of determination of the equation with values close to 1.0 indicating a reliable data fit. 

Examples

Millstone Nuclear Power Plant

The Millstone Nuclear Plant on Long Island Sound in Connecticut installed VFDs on the pumps. A report published prior to the planned operating start date of spring 2011 indicated that the goal was to reduce impingement mortality and entrainment of winter flounder which are present during the period of April to May. The plant has agreed with the NPDES Permit authority to reduce their 2.2 billion gallons per day (bgd) intake flow by 40% during the period of increased biological activity. Flow reduction will be required from April 4 to June 5 or until the source water reaches 52 [o]F, whichever happens first. 

To continue operating normally, if the facility is required to reduce its intake flow, it will need to discharge water at a higher temperature. To facilitate this, the new NPDES permit allows for an increase in the discharge delta T limit for this period. Table 3 below shows the revisions in permit delta T limits and the associated flow reductions if the plant was operating at the limit. Table 2 shows that the increase in the delta T limit would result in flow reductions of 30% and 26% for units 2 and 3 respectively, if the units are operated with delta T values close to the limits. To ground-truth the viability of the flow reduction goal of 40% from April 4 to June 5, a review of NPDES discharge data, as reported in the Permit Compliance System (PCS) (EPA 2008) was conducted. This review indicates why the calculated reductions based on the delta T limits are lower than the proposed reduction goal of 40%. PCS data indicate that the reported actual monthly maximum delta T during a typical April to May period was typically 20-25 degrees, which is much lower than the limit. If it is assumed that the existing delta T is 23 [o]F during this period (selected from within the range of observed values), the corresponding flow reduction would be 44% if the operating discharge delta T was close to the limit of 41 [o]F. Since VFD control systems using real-time temperature sensors can be programmed to reliably control the pump flow such that the discharge temperature may be close to but not exceed the NPDES limitations, a 40% reduction appears to be achievable.

                                   Table 3.
               Millstone Nuclear Plant NPDES Temperature Limits 
                    And Calculated Percent Flow Reductions
Millstone Nuclear
                             Normal Delta T Limit
                        Reported Spring Max Delta T[a]
                                 Seasonal VSP 
                                 Delta T Limit
            Calculated Flow Reduction Based on Change in Limits[b]
        Calculated Flow Reduction Based on Reported Max & Limits[c]
 
                                     [o]F
                                     [o]F
                                     [o]F
                                       %
                                       %
Unit 2 Condenser Discharge
                                      32
                                       -
                                      46
                                      30%
                                       -
Unit 3 Condenser
                                      28
                                       -
                                      38
                                      26%
                                       -
Combined Discharge (Revised Limits)
                                      32
                                      23
                                      41
                                      22%
                                      44%
Combined Discharge (Old Limits)
                                      32
                                      23
                                      32
                                      0%
                                      28%
[a] Typical spring maximum delta T from PCS data.
b  Calculated achievable flow reduction if operating delta T increases from Normal to Seasonal limit value.
c  Calculated achievable flow reduction if operating delta T increases from Reported spring max  to seasonal limit value.

For this base load plant, it appears that a reduction of about 25% could be achieved even without the modification to the discharge limits and that the modification increases the reduction to about 40%. However, the achievable reduction may be lower in warmer months (e.g., June  -  September) regardless of the delta T limit due to the need to maintain the discharge temperature below the maximum limit which in this case was 105 [o]F. PCS data suggests that during peak summer temperatures the delta T would need to be maintained below around 29 [o]F. An increase in delta T from 23 [o]F to 29 [o]F would result in a flow reduction of 20%.

This example shows that significant flow reduction can be achieved even at base load plants, especially during cooler months when high discharge temperature limits do not limit achievable reductions. However, flow reduction (especially during periods when the intake water is warmer) can result in a turbine efficiency loss which presumably is why the permit flow reduction requirement for Millstone is limited to the optimal spawning season for winter flounder of April-May. The NPDES permit fact sheet mentions studying reductions at other times and, as suggested above, since they operate well below the maximum delta T for most of the year, there is an opportunity for substantial flow reduction even under less optimal conditions.

Indian Point

Unit 3 at the Indian Point Nuclear Plant uses VSPs. Table 4 presents the overall monthly flow reductions attained from baseline that were presented in a technology evaluation report. (Enercon Services 2010) The recommended lower limit for tube velocity of 3 fps generally limits the flow reduction to around 50%.

                                   Table 4.
      Variable Speed Pump Flow Reduction Reported for Indian Point Unit 3

                              Month Baseline Flow
                                     (mgd)
                            Historic Operating Flow
                                     (mgd)
                                       
                                    Average
                                Flow Reduction
January
                                    1261.4
                                     600.1
                                     52.4%
February
                                    1261.4
                                     583.2
                                     53.8%
March
                                    1261.4
                                     500.9
                                     60.3%
April
                                    1261.4
                                     614.8
                                     51.3%
May
                                    1261.4
                                     899.6
                                     28.7%
June
                                    1261.4
                                    1199.1
                                     4.9%
July
                                    1261.4
                                    1227.1
                                     2.7%
August
                                    1261.4
                                    1217.4
                                     3.5%
September
                                    1261.4
                                    1234.8
                                     2.1%
October
                                    1261.4
                                    1170.1
                                     7.2%
November
                                    1261.4
                                     806.5
                                     36.1%
December
                                    1261.4
                                     715.2
                                     43.3%
Average Annual
                                    1261.4
                                     899.0
                                     28.7%
           mgd  -  Million gallons per day.
           that were rd to follow. reword.w and tables above? is in deg F?e table to develop an equation which this calculation comes from?Source: Enercon 2010. Table 3.2.
           

This example shows that flow reductions at facilities with variable speed pumps can have considerable seasonal variations. The report notes that the design, efficiency, and operational requirements of the condensers and low pressure turbines affect the flow levels but only specifically mentions the lower limit for condenser tube velocity which limits the maximum reduction achievable if no other limitations are in effect. Although other operational limitations were not specified in this context, factors contributing to the smaller reductions during the warmer months may include: the limiting effect of NPDES temperature limitations similar to the Millstone example above; limits regarding the maximum allowable turbine back pressure which can effectively establish a maximum condensing temperature; and self-imposed limit concerning the acceptable turbine efficiency penalty. The Unit 3 low vacuum alarm is set at no higher than approximately 4 In Hg which is equal to a maximum condensing temperature of 123 [o]F based on data in Table 2. However, the NPDES maximum discharge temperature is 110 [o]F (93.2 [o]F between April 15 and June 30) which is much lower than 123 [o]F, indicating that the NPDES limits, and not the maximum turbine back pressure, are the limiting factors that determine the maximum achievable reductions during warmer months. The fairly low reduction of several percent that occur during the summer months suggest that the self-imposed limit concerning the acceptable turbine efficiency penalty may also be an important factor. As can be seen in Table 2, as the condensing temperature and corresponding turbine back pressure increase, the turbine efficiency penalty increases in an exponential manner and the resulting lost generation can have a significant economic and regional power supply impact that must be considered.
References

ASA Analysis & Communication, Inc. Design and Construction Technology Review
for the Port Jefferson Power Station. August 2007. DCN 12-6815.

Electric Power Research Institute (EPRI). Closed-Cycle Cooling System Retrofit Study Capital and Performance Cost Estimates. Technical Report Number 1022491. 2011. DCN 10-6951.

Enercon Services, Inc. Evaluation of Alternative Intake Technologies at Indian Point Units 2 & 3. February 12, 2010. DCN 10-6962.

EPA. Envirofacts PCS-ISIS Search. Data for NPDES CT0003263.
Accessed at: http://www.epa.gov/enviro/facts/pcs-icis/search.html

Gambica Association Ltd. Variable Speed Drive Pumps  -  Best Practice Guide. 
www.gambica.org.uk/pdfs/VSD_Pumps.pdf. DCN 12-6813

Heat Exchange Institute. Standards for Steam Surface Condensers. 10th Edition. 2006.

Janikowski, D. S. Plymouth Tube Co. Selecting Tubing Materials for Power Generation Heat Exchangers. Presented at the Southwest Chemistry Workshop (Irving, TX): July 28-31, 2003. DCN 12-6818.

Lee&Ro Inc. "Project Description for Pacheco Pumping Plant" Accessed at http://www.lee-ro.com/Proj-Electrical.htm. DCN 12-6814.

MIRANT DELTA, LLC  -  DRAFT. Mirant Proposed BDCP Covered Activities. 2007. DCN 12-6817.

PG&E National Energy Group. Brayton Point Station Permit Renewal Application NPDES Permit No. MA0003654. 316(a) and (b) Demonstration. Volume 4 Section 3.5.3 Variable Speed Drive Pumps. November 2001. DCN 12-6811.

Pacific Gas & Electric (PG&E). Reexamination of Alternatives to Reduce Losses of Striped Bass at the Contra Costa and Pittsburg Power Plants. December 9, 1985. 
Accessed at: http://www.calwater.ca.gov/Admin_Record/C-048529.pdf. DCN 12-6816.

Poseidon Resources for San Diego Regional Water Quality Control Board. Carlsbad Seawater Desalination Project  -  Flow Entrainment and Impingement Minimization Plan. Chapter 4. March 6, 2008. Accessed at: http://www.slc.ca.gov/Reports/Carlsbad_Desalination_Plant_Response.html. DCN 12-6812.



