Hydrocarbon Processing - June 2021 - 30

Process Optimization
In search of the root cause. The bottom
seal pan arrangement is shown in
FIG. 2. The bottom downcomer enters a
242-mm-wide primary seal pan, clearing
its floor by 67 mm. Liquid from the primary
seal pan overflows a 117-mm-tall
overflow weir into the 386-mm-wide secondary
seal pan. From the secondary seal
pan, liquid descends onto the top shed
deck via a 4-in.-diameter, 800-mm-long
pipe. The overflow weir of the secondary
seal pan rises 65 mm above the overflow
weir of the primary seal pan, so no liquid
should be overflowing the secondary seal
pan unless the 4-in. pipe plugs. FIG. 2 also
shows the 10-in. crude feed nozzle at an
elevation about two-thirds of the way
down along the 4-in. pipe.
When the tower was inspected during
the 2018 turnaround, the 4-in. pipe was
found broken and sitting on the top shed
deck. It broke where it exited the secondary
seal pan, leaving a crack and a 4-in. hole
in the seal pan floor. The seal pan floor
was repaired, and the pipe was reinstated.
During the next turnaround in 2020, the
tower was inspected and, once again, the
4-in. pipe was found broken at the same
spot, leaving a crack and a 4-in. hole in the
secondary seal pan floor (FIG. 3).
FIG. 4 shows the cause of the pipe
breakage. It was in front of the flashing
crude feed inlet, and the high feed velocities
caused the pipe to vibrate and break.
This breakage was linked to the entrainment
experience in the tower. With the
pipe broken, the high-velocity vapor was
theorized to pick up the liquid descending
from the secondary seal pan and entrain
it up the tower, causing the naphtha
to turn dark. A solution was sought to
either relocate the 4-in. pipe or to extend
the 10-in. inlet pipe-with baffling or a
pipe distributor-to prevent impingement
of the feed on the 4-in. pipe.
Bazan invited Fluor to join their task
force to seek out the root cause of the
problem. The joint task force review of
FIG. 2. Primary and secondary seal pan
arrangement showing the 4-in. down pipe from
the secondary seal pan to the top shed deck.
the past experiences questioned the theory
that entrainment of liquid falling from
the secondary seal pan is capable of being
the root cause for the dark naphtha. The
liquid flowrate in the upper trays of preflash
towers is small-in some cases, too
small to keep the downcomers sealed. A
4-in. pipe does not transport much liquid.
As long as the downcomers do not lose
their seals, they will drain the liquid entrained
from the broken 4-in. pipe. This
small amount did not explain the flooding
indicated by the observed pressure drop
rise and its persistence for 12 hr. It would
take a massive amount of liquid to explain
the pressure drop rise and its staying elevated
for 12 hr, and this can only come
from the incoming crude.
Another important observation is that
the upper seal pan was not damaged, so the
bottom downcomer will always remain
sealed. This would permit the bottom tray
to continue operating normally. Since the
liquid flow rate in the rectifying trays is
small, any entrainment of this liquid will
be captured by the bottom tray and will
descend via the upper seal pan into the
lower seal pan (FIG. 2). It will then try to
come out through the hole left by the broken
pipe. If the rising vapor prevents this
liquid from descending through the hole,
it will overflow the weir on the secondary
seal pan and, again, flooding will not occur.
The task force identified two mechanisms
by which a massive carryover of
crude is likely to occur. These include:
1. Preflash towers that often
experience foaming1,2
(a foam-over
2. High ρM
usually causes massive carryover
of crude to the upper section)
2 (where ρM
VM
is the
mixed phase density, kg/m3
VM is the mixed phase velocity
at the tower entrance, m/sec).
; and
Both mechanisms are consistent with
the observed dark naphtha, high pressure
drop, and the drop in bottoms level.
Foaming may be caused by components
in the crude1
, desalter chemicals2
or pockets
of water that quickly vaporize upon
entering the tower.
A weak argument against the foaming
theory is the stability of the tower bottom
level. In many cases where preflash foaming
occurs, the bottom pump cavitates
and the bottom level is unsteady. Neither
was observed in the Bazan preflash tower.
The high inlet velocities and foamover
mechanisms are interconnected, as
has been demonstrated by Barber and
Wijn, who point out the significant influence
of the feed inlet design on preflash
foaming.1
A common practice to cure or
prevent foam-overs in preflash drums,
towers and similar services is to pass the
feed through vortex clusters.3,4
With the crude unit coming out of turnaround,
adding vortex clusters was not an
option. A quick solution was sought.
FIG. 3. Broken 4-in. down pipe and the 4-in.
hole in the secondary seal pan floor.
30 JUNE 2021 | HydrocarbonProcessing.com
FIG. 4. The broken 4-in. down pipe hand-held
in its location in front of the tower crude feed
inlet pipe.
Hydraulic analysis and root cause.
The flashing crude feed entered the tower
via a 10-in. pipe, as seen in FIG. 4. The
bottom of the 10-in. crude inlet pipe was
http://www.HydrocarbonProcessing.com
Hydrocarbon Processing - June 2021

Table of Contents for the Digital Edition of Hydrocarbon Processing - June 2021
