Proceedings of the ASME 2014 Power Conference

POWER2014
July 28-31, 2014, Baltimore, Maryland, USA

POWER2014-32177

PERFORMANCE CORRECTIONS FOR STEAM TURBINES WITH MULTI-PRESSURE CONDENSERS

Thomas P. Winterberger Ioannis Tzagkarakis

Senior Technical Manager Performance Engineer
General Electric Power & Water General Electric Power & Water
Schenectady, NY, USA Glasgow, UK

ABSTRACT model the performance of the turbine cycle when multi-

pressure operation exists.
An accurate correction methodology is essential when
analyzing test data and trending performance. One of the
most critical parameters for steam turbines, which can result INTRODUCTION
in large corrections, is condenser back-pressure or exhaust
pressure. ASME Performance Test Code 6 provides guidance on
allowable deviations in test conditions to ensure a good test is
Many large fossil and all nuclear steam turbines are run and to minimize the magnitude of corrections. While
configured with either two or three low pressure condensing guidance is given on allowable fluctuations in exhaust
exhaust hoods. All exhaust hoods may not operate with the pressure and exhaust pressure is included as a Group 2
same condenser pressure. Factors affecting condenser correction, clear guidance is not given on how turbines with
pressures between hoods could include the cooling water multi-condenser shells operating at different pressures should
arrangement, uneven fouling, tube plugging, air removal be handled. This paper will introduce factors which drive
effectiveness, etc. The biggest impact is likely due to the differences in exhaust pressure between condenser shells.
cooling water arrangement. In a parallel arrangement, the While this is primarily associated with a series cooling water
condenser cooling water splits between the shells with each path, other mechanisms can drive differences in the operating
shell receiving an equal amount of flow at the same inlet pressures. These mechanisms, including an estimation of
temperature. In a series arrangement, all cooling water enters their impact on condenser performance, will be presented.
and exits the first shell before entering the next shell. In this
arrangement the cooling water temperature entering the The main discussion of this paper will focus on how error can
second shell is higher than the temperature entering the first be introduced in the calculation of corrected performance if
shell resulting in the condensers operating at different exhaust differences in pressures between condenser shells are not
pressures. properly accounted for. Recommendations will be given on
how to properly create exhaust pressure corrections for this
One common practice is to apply a single exhaust pressure situation and how to properly apply them to avoid the
correction factor based on the average exhaust pressure of all introduction of any error associated with the correction
condenser shells. In cases where the differences in condenser methodology. Recommendations will also be given on how
pressure are small, this practice can provide accurate to properly analyze and trouble-shoot exhaust pressure data
corrected turbine performance. As the difference in for multi-pressure operation.
condenser pressures increases, the potential for introducing
error in the corrected performance results also increases.
NOMENCLATURE
This paper will discuss the mechanism of why multi-pressure
operation can result in correction errors if not modelled HP - High Pressure
correctly and will also quantify the potential impact of these IP - Intermediate Pressure
errors on the corrected performance results. In addition,
guidance will be given on how exhaust pressure correction LP - Low Pressure
curves should be created and applied to most accurately LSB - Last Stage Bucket

1 Copyright 2014 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 11/25/2014 Terms of Use: http://asme.org/terms

 OEM - Original Equipment Manufacturer plugging, uneven cooling water flow distribution between the
shells etc. These factors will be discussed in Section 2.
PTC - Performance Test Code
A possible configuration where we encounter a Series cooling
ST - Steam Turbine
water path is shown in Figure 1.3. In this example each
VAN - Annulus Velocity condenser shell has its own hotwell drain which would
combine prior to the condensate pumps.
A variation of the above design is the Cascading Hotwell
1) CONDENSER CONFIGURATIONS configuration, shown in Figure 1.4. In this case, the
Many different condenser configurations exist to meet the condensate stream cascades from the Low Pressure shell into
requirements of each individual application. The simplest the High Pressure shell internally before being collected in a
configuration would be a single-shell condenser with either common hotwell.
one or two passes. For this configuration performance
A multiple outlet hotwell design can be seen in either a series
calculations are relatively straightforward as there is a single
or parallel cooling water path configuration, while a single
operating pressure in the single-shell. outlet hotwell is more common in series cooling water path
Larger steam turbine units are normally configured with two configurations.
low pressure exhaust hoods with each LP hood exhausting to
While this paper will focus on configurations with two
its own condenser shell. The two condenser shells could be
condenser shells, very large fossil units and many nuclear
configured with either a series or a parallel cooling water
power plants are configured with three LP exhaust hoods
flow path as shown in Figures 1.1 and 1.2 respectively. feeding three condenser shells. These configurations can
utilize either series or parallel cooling water paths.

2) MECHANISMS AFFECTING CONDENSER

PERFORMANCE AND EXHAUST PRESSURE

Condenser
There are several mechanisms that can result in condenser
Condenser
ShellA ShellB performance deterioration which drive condenser pressures to
be higher than expected. These mechanisms may include air
removal effectiveness, tube fouling and reduced cooling
water flow. These mechanisms primarily impact heat transfer
effectiveness through either increased resistance, reduced
surface area, or a combination of both.
Figure 1.1Series Cooling Water Path Air accumulation in the condenser reduces the heat transfer
rate and is caused by either reduced vent capacity or too high
of an air ingress rate. In the first case, the low vent rate is
related to the air removal effectiveness and the associated
equipment used (Vacuum Pump or Steam Jet Air Ejector),
while in the latter case the increased ingress rate is caused by
excessive air leakage into the condenser system.
Condenser Condenser
ShellA ShellB
Tube fouling typically occurs on the tube side (Cooling Water
side) by the accumulation of foreign material (Biological,
chemical, etc) on the inside diameter surfaces. This increases
heat transfer resistance from tube side to shell side, resulting
in the condenser operating at an elevated pressure.
Tube plugging is caused by the accumulation of debris and
Figure 1.2Parallel Cooling Water Path results in blocking the flow to some of the tubes and therefore
reducing the active heat-transfer surface area of the
condenser. Tubes could also be intentionally plugged to
In the case of a series cooling water path the cooling water
mitigate tube leaks.
enters and exits Shell A (Low Pressure) before entering Shell
B (High pressure). The increased temperature of the cooling Reduced Cooling Water flow rate is typically caused by a
water entering Shell B is the main factor contributing to a reduction in pump capacity. Significant amounts of tube
higher pressure in Shell B. In the case of a parallel cooling plugging can impact pump capacity by a reduction in flow as
water path, the temperature of the cooling water at the inlet of a consequence to a corresponding increase in discharge head.
each Shell is equal, resulting in similar shell pressures. For a given condenser duty, reduced cooling water flow rate
However, the resulting pressure in each shell in a parallel typically results in an elevated temperature difference across
cooling water path may again differ due to other factors such the water box.
as shell-side fouling, cooling water tube fouling and
Additional factors to consider are properly filled water boxes
and properly installed pass rib gaskets for multi-pass water

2 Copyright 2014 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 11/25/2014 Terms of Use: http://asme.org/terms

 Figure 1.3Two Condenser Shells with Individual Hotwell Outlets

Figure 1.4Two Condenser Shells with Single Hotwell Outlet

box designs. A water box that is not properly vented can pass causing an elevated condenser pressure. A lower than
result in a significant amount of air trapped at the top of the expected water box differential pressure, a lower than
water box. This trapped air can blanket the upper tubes and anticipated water box differential temperature and a lower
reduce the effective surface area. Typically the symptom of than expected outlet temperature for a given inlet temperature
an insufficiently vented water box is an elevated temperature and condenser duty are possible symptoms of a dislodged or
difference across the water box for a given condenser duty missing water box pass rib gasket.
along with a lower than expected water box differential
All the factors mentioned above may result in higher than
pressure.
expected condenser shell pressures. Table 1.1 gives
On multi-pass water box designs, a pass rib gasket prevents approximate sensitivities of several key condenser design and
cooling water flow from bypassing one water box chamber to operating variables on condenser pressure. The data from
another. A damaged or missing gasket would result in Table 1.1 shows that cooling water inlet temperature has the
reduced cooling water flow through a particular tube bundle largest impact on condenser pressure indicating that the
cooling water arrangement (series vs parallel) is the biggest
Approximate factor in driving pressure differences between condenser
Mechanism Impact on Con- shells.
denser Pressure

10% Plugged Tubes 0.010.1 HgA 3) IMPACT OF EXHAUST PRESSURE ON TURBINE

PERFORMANCE
10% Change in Cooling Water Flow 0.010.2 HgA
Figure 3.1 shows a schematic of the turbine-condenser
10% Change in Percent Clean 0.010.3 HgA interface for a typical large steam turbine configuration. For
the purposes of this paper all examples will reference back to
10 F Change in Cooling Water Inlet
0.201.0 HgA this configuration which is typical for a 550 MW large fossil
Temperature
steam turbine application. This configuration has two LP
Table 1.1Condenser Impacts on Operating hoods with a total of four LP steam flow paths. Each LP
Pressure hood exhausts to its own condenser. The velocities of each

3 Copyright 2014 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 11/25/2014 Terms of Use: http://asme.org/terms

 LP exhaust flow will be dictated by the operating pressure of shape of the exhaust loss curve, the average of the losses at
the condenser it is feeding. each velocity may not be equal to the loss at the average
velocity. The example in Table 3.1 illustrates this point for a
Changes in exhaust pressure affect turbine performance in
case where the LP condenser is at 1.5 HgA and the HP
two ways. Annulus velocity changes impact total exhaust
condenser is at 3.5 HgA. The exhaust loss at either pressure
loss and LP turbine efficiency. In addition the energy
is much higher than the exhaust loss corresponding to VAN
available to do work changes with changes in exhaust
at 2.5 HgA which is the average of the LP and HP condenser
pressure. In absolute pressure terms, lower exhaust pressure
pressures.
results in more available energy while higher exhaust
pressures result in less available energy. With all else being 3.2) Exhaust Pressure Correction Curve
equal, annulus velocity increases as exhaust pressure
decreases. P exh VAN Exh Loss
3.1) Exhaust Loss Curve HgA ft/sec Multiplier

Each last stage bucket (LSB) and exhaust hood configuration Cond 1 1.5 870 2.9
is characterized by an exhaust loss curve. The exhaust loss
Cond 2 3.5 400 2.6
includes the effects of loss due to steam velocity not
converted to work, pressure drop through the exhaust hood, Cond Avg 2.5 540 1.1
and inefficiencies related to LP stage performance with low
velocities. Figure 3.2 shows a typical exhaust loss curve as a Table 3.1Exhaust Loss Calculation
function of annulus velocity leaving the last stage. For most
last stage bucket and hood configurations, the optimum point
In order to correct measured turbine performance to a set of
with minimum exhaust loss occurs around 600 ft/s. Below
reference conditions in accordance with ASME PTC 6
600 ft/s results in operating in the turn-up region where
methodology, Group 1 and Group 2 correction curves
lower velocities can result in flow recirculation leading to
supplied by the turbine OEM are used. Correction curves for
increased losses. Above 600 ft/s results in operation in a
exhaust pressure are included in the Group 2 corrections. The
region dominated by hood loss and leaving loss that increase
curve is generated using a heat balance modeling program
with increasing annulus velocity
and takes into account exhaust loss, changes in available
Annulus velocity (VAN) is a function of annulus area, flow, energy and turbine efficiency, and moisture effects. Figure
steam quality (moisture), and specific volume. Specific 3.3 shows an example exhaust pressure correction curve for a
volume increases exponentially with decreasing exhaust large steam turbine configuration operating at a single
pressure so small changes in pressure can result in large condenser pressure. The curve shows how output changes
changes in velocity. For a unit operating with multiple flow with variations in exhaust pressure. This curve could also be
ends, there is assumed to be equal flow leaving all turbine generated to show heat rate variation with exhaust pressure.
ends. With this assumption, turbines operating with multiple A single curve is shown on Figure 3.3 representing full load
condenser pressures will operate with different annulus throttle flow while an actual curve would cover a wider range
velocities entering each condenser shell. Due to the parabolic of expected operating flows. This curve could be applied to

Figure 3.1Turbine-Condenser Schematic for Typical Large Steam Turbine Configuration

4 Copyright 2014 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 11/25/2014 Terms of Use: http://asme.org/terms

 15

Exhaust Loss Multiplier

10

0
0 200 400 600 800 1000 1200 1400 1600
Annulus Velocity (ft/sec)

Figure 3.2Typical Steam Turbine Exhaust Loss Curve

4.0% 2.0%
LP Condenser
3.0% 1.5%
HP Condenser

2.0% 1.0%

1.0% 0.5%
% Change in Output
% Change in Output

0.0% 0.0%

-1.0% -0.5%

-2.0% -1.0%

-3.0% -1.5%

-4.0% -2.0%

-5.0% -2.5%

-6.0% -3.0%
0.0 1.0 2.0 3.0 4.0 5.0 6.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0

Exhaust Pressure (in HgA) Exhaust Pressure (in HgA)

Figure 3.3Typical Steam Turbine Exhaust Figure 3.4Typical Steam Turbine Exhaust
Pressure Correction Curve for Single Pressure Pressure Correction Curve for Multi-Pressure
Operation Operation

5 Copyright 2014 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 11/25/2014 Terms of Use: http://asme.org/terms

 turbines operating with a parallel flow condenser assuming HP Condenser: 1.0 to 5.0 HgA
both condenser shells are operating at the same pressure. Minimum Difference 0.5 HgA
Figure 3.4 shows an exhaust pressure correction curve for a Maximum Difference 4.0 HgA
turbine operating with multiple condenser pressures. The
HP pressure is always greater than LP pressure
curve gives lines for both the low pressure and high pressure
condensers. The magnitude of the correction for each of 33.5 and 26 LSB
these curves is approximately one half the magnitude of the The case studies were run for a nominal 550 MW unit with
curve shown in Figure 3.3. These actual curves would apply steam conditions of 2400 PSIG / 1000 F / 1000 F. Each
to the configuration shown in Figure 1.4 where there is a GateCycleTM run was corrected back to a reference exhaust
single hotwell outlet. Since the HP condenser alone is setting pressure. Figure 3.4 was applied first using the individual
the hotwell outlet temperature it has a slightly larger pressures for the HP and LP condensers and then Figure 3.3
influence on the overall turbine-cycle performance. was applied based on the average pressure between the HP
and LP condensers. The difference between the Figure 3.4
4) CASE STUDIES corrected output and the Figure 3.3 corrected output is the
error introduced when trying to correct based on the average
The GateCycleTM heat balance modeling program was used to pressure.
simulate test and operating conditions to quantify the
potential error that can be introduced if exhaust pressure 4.1) Case Study 1: 33.5 Last Stage Buckets
corrections are not applied properly. A matrix of runs were For the example conditions used, a 33.5 LSB would be more
completed with the following constraints: optimum for a lower backpressure application in the range of
LP Condenser: 1.0 to 5.0 HgA 1.52.0 HgA. In this backpressure range VAN is 650

HP ("HgA) 33.5" Last Stage Buckets

5.0 -0.79% -0.10% 0.26% 0.35% 0.29% 0.16% 0.07% 0.02% 0.00%
4.5 -0.94% -0.26% 0.11% 0.24% 0.20% 0.09% 0.03% 0.00%
4.0 -0.99% -0.38% -0.02% 0.12% 0.11% 0.04% 0.00%
3.5 -0.96% -0.40% -0.10% 0.03% 0.04% 0.00%
3.0 -0.78% -0.31% -0.07% 0.01% 0.00%
2.5 -0.55% -0.17% -0.02% 0.00%
2.0 -0.28% -0.06% 0.00%
1.5 -0.04% 0.00%
1.0 0.00%
LP ("HgA) 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Table 4.1 - Error in Corrected Performance Calculations Case Study 1

HP ("HgA) 26" Last Stage Buckets

5.0 -1.21% -0.77% -0.40% -0.16% -0.05% 0.00% 0.03% 0.02% 0.00%
4.5 -1.10% -0.69% -0.35% -0.14% -0.04% -0.01% 0.01% 0.00%
4.0 -0.94% -0.58% -0.27% -0.09% -0.01% 0.01% 0.00%
3.5 -0.74% -0.45% -0.20% -0.05% 0.00% 0.00%
3.0 -0.52% -0.30% -0.11% -0.02% 0.00%
2.5 -0.30% -0.14% -0.03% 0.00%
2.0 -0.12% -0.04% 0.00%
1.5 0.08% 0.00%
1.0 0.00%
LP ("HgA) 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Table 4.2 - Error in Corrected Performance Calculations Case Study 2

6 Copyright 2014 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 11/25/2014 Terms of Use: http://asme.org/terms

 700 ft/s which is near the minimum point on the exhaust loss velocities. This case was run to determine if the magnitude of
curve. Table 4.1 shows the results of the 33.5 LSB the correction error varied with the LP exhaust configuration.
GateCycleTM runs in terms of correction error. A negative
The results of the Case Study 2 runs are shown in Table 4.2.
number indicates that corrected performance is being under-
While the magnitude of the difference between the highest
estimated when the correction is based on the average of the
and lower errors appears to be very close between the two
HP and LP condenser pressures. A positive number indicates
case studies, the negative errors increased in magnitude while
the opposite with corrected performance being over-
the positive errors decreased. This was expected as the
estimated. Table 4.3 shows a sample calculation of how the
optimum exhaust pressure for the 26 LSB is closer to the
error in the correction methodology is determined where
high-end of the analyzed operating range and further away
corrected output is calculated as follows:
from the low end of the operating range. Again the error does
KWCORR = KWTEST not appear to become significant until the difference in
pressures between the condensers exceeds 0.5 HgA.
(1-Correction Curve Reading)
5) OTHER CONSIDERATIONS
Analysis of the data in Table 4.1 indicates that significant
error is possible and that the magnitude tends to be higher at In addition to the topics previously discussed there are several
lower LP pressures and increases as the difference between other key areas related to exhaust pressure where attention
the HP and LP pressures increases. The error in corrected needs to be paid to ensure an accurate and consistent test
performance does not appear to start being significant until program. These include instrumentation, data analysis, and
the difference in pressures exceeds 0.5 HgA. how to properly create exhaust pressure correction curves.
5.1)Instrumenta on
Simulated Test Condition with GateCycle
Due to spatial variations within any given condenser shell, it
LP Condenser 1.0 HgA is recommended to measure exhaust pressure at multiple
HP Condenser 3.5 HgA
locations with basket tip pressure connections. Large steam
turbines supplied by GE are typically equipped with four
Predicted Output 546,224 kW basket tip connections per exhaust end for a total of eight per
exhaust hood. The two hood configuration shown in Figure
Output Corrected for LP and HP Pressures 3.1 would be equipped with a total of sixteen basket tip
Figure 3.4 LP Correction 1.0160 exhaust pressure connections.

Figure 3.4 HP Correction 0.9932 With any reading, measurement errors can contribute to
errors in calculated results. As indicated in ASME PTC 6,
Corrected Output 541,303 discrepancies in excess of 0.1 HgA in redundant exhaust
pressure readings should be investigated further. Two
Output Corrected for Average Pressure common issues related to exhaust pressure measurement
Average of HP and LP errors are damaged/missing basket tips or moisture
2.25 HgA accumulation in the sensing lines. Plants are encouraged to
Condenser Pressures
inspect basket-tip connections during outages to verify that
Figure 3.3 Correction at
1.0189 they are intact and undamaged. Horizontal runs in the
Average Pressure
sensing lines should be avoided as they can result in moisture
Corrected Output 536,092 build-up. Its good practice to open the transmitter vent and
allow the condenser vacuum to pull the moisture out of the
Error in Corrected Output sensing line prior to testing/assessment activities as long as
% Difference -0.98% the transmitter is not a turbine trip sensor.
The installation of the basket tips (the exact position and
Table 4.3Sample Calculation of Corrected angle of the basket tip to the flow) may vary from one
Output Error
condenser shell to another. This can introduce small
differences in the measured pressure between two shells.
4.2) Case Study 2: 26 Last Stage Buckets Local variations of the flow field may also contribute to
erroneous exhaust pressure measurements. If all basket-tips
If the turbine cycle modeled in Case Study 1 was expected to cannot be utilized for test purposes, judgment should be used
operate with higher condenser pressures in the range of 2.5 - to select a mix of Left/Right and Corner/Middle basket-tips in
3.0 HgA, a 26 LSB might be the more optimum choice. In an attempt to capture flow field variations.
this backpressure range, VAN is in the range of 650700 ft/s
and again results in operation close to the minimum point on 5.2) Data Evaluation, Analysis and Trending
the exhaust loss curve. As condenser pressure increases, Table 3.1 of ASME PTC 6 specifies criteria for the allowable
specific volume and volume flow decrease. As such less deviations of test average conditions from reference
annulus area is required to maintain the same annulus conditions. For exhaust pressure the allowable deviation is +/
- 0.05 psi or +/- 2.5% of the absolute pressure, or whichever

7 Copyright 2014 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 11/25/2014 Terms of Use: http://asme.org/terms

 is larger. When dealing with multiple condenser shells curvature.
operating at different pressures, the criteria should be applied
to each individual condenser shell. In addition the criteria in Hold the HP condenser fixed at the top end of the
Section 5.1 above relating to 0.1 HgA discrepancies in expected operating range and vary the LP condenser
redundant readings should be applied to all readings within pressure across its expected operating range with a
the same condenser shell. sufficient number of points to produce a smooth curve.
Additional points may be needed in areas with more
As has been discussed, an accurate correction methodology is curvature.
critical when conducting performance testing to determine
contractual acceptance. The proper application of exhaust Hold BFPT exhaust pressure fixed when varying main
pressure corrections is also critical for long-term trending of condenser pressures. This isolates BFPT performance
performance data. Large changes in condenser pressure can from the main turbine and allows BFPT performance
be seen due to seasonal variations in ambient conditions. to be accounted for with its own correction curves.
Even if the relative pressure between condenser shells Drains and other heat loads terminating in the
remains constant but the absolute values of the pressures condenser should also be held constant.
change, the magnitude of the error due to not properly
correcting can vary significantly. Based on the data in Table
4.1, the error associated with a 2.0 HgA difference in CONCLUSION
pressure between condenser shells can vary from 0.78% to Condensers, like turbines, can be configured in a number of
+0.29%. ways. Many condensers for large turbine applications are
A common calculation for analyzing exhaust pressure configured with a series cooling water path. This results in
measurements is to compare the calculated saturation the condenser shells operating with different cooling water
temperature at the measured exhaust pressure to the measured inlet temperatures and thus different pressures. Other factors
hotwell outlet temperature. This would be done both for for both parallel and series condensers can drive differences
troubleshooting purposes and also to calculate condenser sub- in operating pressures including tube plugging, cooling water
cooling to determine if a Group 1 sub-cooling correction flow differences, and heat transfer degradation. When a
needs to be made. When making this comparison it is difference in pressure does exist care must be taken when
important to know if each condenser shell has its own hotwell correcting turbine performance whether it is being done for
outlet or if there is a single hotwell outlet coming just from formal performance testing or long-term performance
the high pressure condenser. The following guidelines should trending. When the difference is below 0.5 HgA, the impact
be used: on performance appears to be minimal and may not need to
be accounted for. Above a 0.5 HgA difference the impact
Single hotwell outlet (Figure 1.4) can become significant exceeding a 1% impact on corrected
Compare measured hotwell outlet temperature to the performance as shown in the Case Studies presented in
saturation temperature calculated from the HP Section 4. The data in Tables 4.1 and 4.2 are not intended to
condenser pressure measurements. be used as test corrections. This data was presented to
highlight the potential significance of the error and to
Individual hotwell outlets (Figure 1.3)
encourage the practice of applying exhaust pressure
Compare measured hotwell outlet temperature for corrections for each individual condenser when differences
each condenser shell to the saturation temperature exist in the condenser shell pressures. While this paper
calculated from the condenser pressure focused on turbine configurations with two LP exhaust hoods,
measurements for that condenser shell. If hotwell the observations and recommendations can also be applied to
outlet temperature is only available after the units with three exhaust hoods.
individual lines have mixed, compare the mixed
temperature to the saturation temperature calculated
based on average of the HP and LP condenser REFERENCES
pressure measurements. ASME PTC 6 2004 Steam Turbines
5.3) Creating Exhaust Pressure Correction Curves
Richard E. Putman: Steam Surface Condensers: Basic
The following are a set of recommendations that can be
Principles, Performance Monitoring, and Maintenance.
followed when creating exhaust pressure correction curves
ASME Press, 2001.
for individual condenser shells:
Configure the model to match the actual plant
configuration as closely as possible including hotwell
outlet configuration and the configuration of the boiler
feed-pump turbine exhausts.
Hold the LP condenser fixed at the bottom end of the
expected operating range and vary the HP condenser
pressure across its expected operating range with a
sufficient number of points to produce a smooth curve.
Additional points may be needed in areas with more

8 Copyright 2014 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 11/25/2014 Terms of Use: http://asme.org/terms