POWER October 2010 - 56

PLANT O&M
2. Thermal turndown. The typical APH
may be as much as 60 feet in diameter. When
the APH rotor is heated from a cold condition
(blue), thermal expansion (yellow) can cause
the rotor to droop or " turn down " up to 3
inches on the periphery. Knowing the amount
of turndown is important when presetting the
seal position before operation, because seal
positions will change as the rotor warms to its
operating temperature. Source: Storm Technologies
Inc.
Hot end of rotor
Leakage path (cold)
Cold end of rotor
Leakage path (hot)
where gas and air bypass the rotor and heat
exchange elements.
Two Types of Seal Leakage
Consider APH leakage as being one of two
types and taking one or more of four distinct
leakage paths through regenerative APH seals
(Paths A, B, C, and D in Figure 3). Each leakage
path affects the economic operation and
heat rate of the plant in a different way:
■ Circumferential/bypass leakage (Paths C
and D) has an effect on heat transfer and
boiler heat rate. A portion of the leakage
Path C will also travel around the outside
circumference of the rotor, joining the
flow of leakage Path A and increasing fan
horsepower requirements.
■ Radial seal leakage (Paths A and B) is
typically calculated as a percentage of the
boiler exit gas flow (Path 2), not as a percentage
of fan input airflow (Path 1).
3. Four leakage paths through
an APH. Path 1 is the normal airflow path
through the APH and Path 2 is the normal flue
gas flow path. Path A shows the forced draft
(FD) fan flow leaking ambient air directly to
the gas outlet duct (through the radial or circumferential
seals). Path B shows preheated
FD fan airflow that short-circuits back to the
gas outlet duct. Path C represents ambient
FD fan air that leaks around the APH and enters
the boiler unheated. Path D represents
hot flue gas exiting the boiler, bypassing the
APH, and exhausting at a high temperature.
Source: Storm Technologies Inc.
Knowing the gas flow leakage (as a percentage)
is convenient when calculating the
" correction " in gas exit temperature that results
from air in-leakage when doing overall
boiler efficiency tests.
Circumferential Seal Leaks. CircumID
fan
ferential seals are located on the entire circumference
of the APH rotor, on both the hot
end and cold end of the APH. On the flue gas
side of the APH, all of the leakage through
the inlet side circumferential seals will shortcircuit
around the APH (bypassing the heat
transfer element) and exit through the downstream
circumferential seals. This leakage
results in a loss of enthalpy transfer into the
element and increases the temperature (and
therefore the actual volume) of gas entering
the induced-draft (ID) fans.
On the air side of the APH, the volume of
FD fan
ends (about 400F). Together, these characteristics
produce significant radial thermal expansion
difference between the hot and cold
sides of the APH's rotor after unit start-up.
It's not uncommon for the outer edges of a
large APH at operating temperature to droop
or " turn down " by 3 inches or more compared
with the cold condition (Figure 2).
This thermal distortion droop opens gaps
in the sealing surfaces that separate the cold
incoming air from the outlet gases as well as
in the sealing surfaces around the circumference
of the APH. Turndown changes the gaps
between both the radial and circumferential
seals and their respective sealing surfaces
56
leakage through the first set of circumferential
seals will enter the annulus around the
perimeter of the rotor, where the leakage will
split in two directions. The volume of leakage
in each direction depends on the differential
pressures between points of exit. A portion
of the flow will continue in a straight path
and exit through the second set of circumferential
seals. The remainder of the flow will
be directed around the perimeter of the rotor
and exit into the exhaust gas stream (through
the axial seals), and that volume will, in turn,
exit the APH through the gas-side cold end
circumferential seals.
Most difficult to account for is air or gas
bypassing the APH around the circumference
of the APH. There is no simple or effective
way to take an actual measurement of leakage
around the circumference of the APH.
These leakage rates must be calculated using
pressure differentials and measuring gaps
www.powermag.com
between the rotor and the circumferential/
bypass seals, with an allowance for structural
deformation at operating temperature, including
rotor expansion and chord distortion
between diaphragms.
One simple estimating approach is to
consider the total leakage cross section as
equivalent to that of a flat plate orifice to
calculate the leakage flow rate. Although it
is not a perfect match to perimeter leakage in
an APH, it provides a closer approximation
than traditional crack flow equations for ventilation
systems, which are more appropriate
for very small openings with very small differential
pressures.
Radial Seal Leaks. Radial seal leakage
represents the percentage increase in outlet
gas flow caused by the mass of inlet air leaking
into the gas outlet stream. (The majority
of this leakage, as measured, is flowing past
the radial seal area, but in reality, this measurement
also includes other leakage paths,
including entrained leakage and axial seal
leakage.) Shockingly, leakage rates attributed
to radial seals have been measured at over
40% in some APHs, and leakage rates around
20% are often accepted as a " normal " condition.
However, this much leakage places a
significant extra burden on the boiler fans in
order to move gas and air that serves no useful
purpose.
In addition, changes in fuels and operating
conditions over the years often push ID fans
to near rated capacity. When a fan is operating
at over 80% of capacity, the slope of
the horsepower/volume curve becomes very
steep. At near full capacity, a 1% increase in
fan volume often results in a 3% increase in
required fan horsepower.
Two Radial Leakage Penalties
Two penalties to boiler performance occur
with excessive radial seal leakage. The first
is the thermal losses associated with the leakage
air cooling the APH. The second is the
additional auxiliary horsepower consumed
by the fans for pushing more flow.
The first step in determining the thermal
loss is to establish the gas outlet temperature
corrected for no leakage. The ASME performance
test code (PTC) assumes that all of
the air in-leakage occurs on the cold side.
However, in reality it is a mix of the hot and
cold side radial leakage. The hot side radial
leakage does not cool the outlet temperature
as much as the cold side leakage does.
Though the exact split will vary and can't
be directly measured, a good assumption is
that the leakage is biased 60/40 to the cold
side due to higher differential pressures and
the higher density of cooler air at the cold
end of the APH. Because the hot side radial
seal leakage returns some of its heat to the
POWER | October 2010
http://www.powermag.com
POWER October 2010

Table of Contents for the Digital Edition of POWER October 2010
