Balance Line
Balance Line
Balance Line
15
2
2
.
Ns
RPMx GPM
TDH
=
3
4
due to the seal flush pressure that is now higher than the crossover
pressure by at least 20 percent. Since all multistage pumps with a
back-to-back impeller arrangement were designed to have fluid
moving toward the stuffing box through the throttle bushing, some
manufacturers only trap the throttle sleeve in one direction. This
can pose a problem when the balance line is closed and the throttle
bushing flow reverses.
Figure 12. Typical Throttle Bushing Design and Leakage Flow
Path on Pumps with a Balance Line.
Figure 13. Throttle Bushing Leakage Flow Path on Pumps without
a Balance Line (Bushing Shown Before Required Modification for
Flow Reversal).
The throttle sleeve (Figures 12 and 13) is the dynamic member
of the throttle annulus and rides inside the throttle bushing. Some
designs use a split axial key to keep the throttle sleeve from
moving. The sleeve is installed on the shaft and has an axial
keyway that mates with the split key. In the balance line closed
case, the flow reversal through the throttle annulus might cause the
throttle sleeve to back off the axial key and severely damage the
pump. Most all designs of this type also incorporate a radial key in
the throttle sleeve to ensure it does not rotate relative to the shaft
during operation. This radial key is typically blind so that high
pressure fluid cannot bypass the throttle annulus through the
keyway. The blind radial key would indeed trap the sleeve in the
opposite direction of the axial key and prevent the throttle sleeve
from moving when the balance line is closed. However, this is not
an acceptable method of axially trapping and locating a sleeve on
a high energy centrifugal pump.
Therefore, if the balance line is to be closed on multistage
pumps of this type, the throttle sleeve must be trapped utilizing an
approved method. This may or may not involve changing the shaft
design of the pump.
There are also manufacturers that use countersunk setscrews or
a threaded sleeve/shaft to secure the throttle sleeve to the shaft.
These designs are inherently bidirectionally trapped and will not
need to be altered for closed balance line operation. The test pump
utilized a set-screwed throttle sleeve and performed acceptably for
the duration of the testing.
Performance Matching
Closing the balance line does have a positive effect on both flow
and developed head. This might present a problem to an end-user
who simply wants to save money on electricity while remaining at
the same rated flow and head as their factory pump. The modified
pump will produce more flow at a given system head and the
customer would not see any noticeable reduction in energy costs.
The modified pump would draw marginally less if not the same
amount of power from the driver as the factory pump in this
condition. To ensure that the modified pump will perform
acceptably in this case, a slight impeller trim will be required. It is
expected that this reduction in impeller diameter of some or all the
impellers will reduce the efficiency gains from closing the balance
line by 25 to 30 percent, depending upon impeller hydraulic design
and magnitude of impeller trim. However, the impeller trim will
enable the end-user to realize some energy savings without system
modification or increased pump throughput.
If the end-user has driver speed variability through a variable
frequency driven motor, steam, or gas turbine, the negative effects
of performance matching can be completely eliminated. Instead of
trimming the impeller to match previous performance of the pump,
the speed of the driver can be reduced slightly until the pump
performs as it did before the balance line was removed. The full
benefit of removing the balance line can then be realized as
reduced electricity, steam, or fuel consumption.
Economic Impacts
Ultimately, cost savings will drive the development and imple-
mentation of this technology. Table 6 is an estimate of the annual
savings versus the cost of modification for pumps similar to the test
unit. Over an estimated 10 year life of the pump, the savings are
considerable. All calculations are based on an energy cost of eight
cents per kilowatt-hour. It is easy to see that a plant with 10 to 20
high-energy pumps would see a significant impact on their energy
bill even on a monthly scale. These are savings that are literally
being thrown away with old technology.
Table 6. Annual Savings and Cost of Modification.
The removal of the balance line in multistage pumps is most
ideally suited to end-users who would like more capacity and/or
discharge pressure out of their existing units. Tables 2 and 3 show
that for a given flow or TDH the modified pump outperforms the
factory condition unit and often at a lower horsepower
requirement. This would be useful for pipeline operators that are
often paid by the barrel of transported product.
CONCLUSIONS
The multistage pump balance line is truly an antiquated feature
on some of todays pumps. High-energy pump efficiency can be
effectively augmented through the elimination of the balance line.
The effects of higher stuffing box pressure, throttle bushing flow
reversal, and axial thrust load can be dealt with in the modification
process. With these issues adequately addressed, there is no reason
pumps without balance lines cannot run as reliably as similar units
with balance lines. Truly, this modification is not ideally suited for
all multistage pumps and care must be taken to properly select
candidate pumps to justify costs. Pending successful extended field
trials and analysis, balance lines can be eliminated from many
multistage centrifugal pumps to increase efficiency and reduce
operating costs.
PROCEEDINGS OF THE TWENTY-FIRST INTERNATIONAL PUMP USERS SYMPOSIUM 2004 93
HP Annual Gross
Savings
Cost of Modification
(less Mech. Seals)
Total Savings Before Overhaul
(Based on 10 Yr Life)
250 $4,834 $8,000 $40,340
500 $9,668 $12,000 $84,680
750 $14,502 $20,000 $125,020
1000 $19,336 $25,000 $168,360
Precautionary Note
The fluid patterns that generate the axial hydraulic thrust in
multistage pumps are highly complex. The pressure distributions
acting on the front shroud and the back shroud of each stage
impeller are heavily influenced by the leakage (amount and
direction, i.e., radially inward or outward). Therefore the
prediction of the hydraulic axial impeller forces requires a suffi-
ciently accurate computer model for the leakage influence.
Moreover, the leakage effect is not yet fully understood. This means
that even more sophisticated theoretical calculations of axial
thrust are determined under several assumptions that require
experimental validation. It is highly recommended that theoretical
predictions of axial hydraulic thrust in multistage pumps,
including configurations with opposite impellers (apparently auto-
balanced), following design changes such as impellers and
bushings, are backed up with actual measurements.
The end-user should always take care, with any modification, to
adequately account for factors that affect pump reliability and
performance. Considerations such as axial thrust, internal flow
reversals, mechanical seal selection, and mechanical seal flush
selection are all very important in performing the modifications
discussed in this paper. If these factors are not specifically
addressed, it is likely that pump reliability will be decreased.
REFERENCES
Karassik, I., Messina, J., Cooper, P., and Heald, C., 2001, Pump
Handbook, Third Edition, New York, New York: McGraw Hill.
Lobanoff, V. S. and Ross, R. R., 1986, Centrifugal Pumps: Design
and Application, Houston, Texas: Gulf Publishing Company.
Miyashiro, H., Lino, T., and Sato, H., 1980, Hydraulic Axial
Thrust in Multistage Centrifugal Pumps, Transactions of the
American Society of Mechanical Engineers, 13, pp. 64-69.
Stephanof, A. J., 1957, Centrifugal and Axial Flow Pumps, New
York, New York: John Wiley and Sons.
MULTISTAGE PUMP EFFICIENCY GAINS THROUGH THE
ELIMINATION OF THE SEAL CHAMBER PRESSURE EQUALIZATION LINE
94