ROLLOVER OF LNG IN TRADING LNG CARRIER - SIGTTO PANEL
pollockenergy
48 views
27 slides
Aug 01, 2024
Slide 1 of 27
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
About This Presentation
Case Study of LNG Rollover
Size: 855.73 KB
Language: en
Added: Aug 01, 2024
Slides: 27 pages
Slide Content
ROLLOVER
(Density-Stratified Liquid Layers)
S.S. DEWA MARU
SIGTTO 59
th
GPC Meeting
SIGTTO 59
th
GPC Meeting
Wednesday 22
nd
April 2009
BOSTON
Presented by: LNG Carrier Group “K” LINE SHIP MANAGEMENT
CO.LTD
and BP SHIPPING LTD
Objectives
R
Brief overview of rollover phenomenon
R
Background to the incident
R
Review of outcomes and pertinent voyage records Review of outcomes and pertinent voyage records during rollover
R
Lessons Learnt
R
Q & A’s
Natural Convection in an LNG
Storage Tank
R
A small amount of boil-off of
LNG is generated by heat
ingress through insulation.
R
Liquid on top layer
evaporates, cools and
becomes denser. This becomes denser. This phenomenon is called
“Weathering”.
R
Natural convection causes
circulation of the LNG within
the tank, maintaining a
uniform liquid composition.
Stratification
o
Stratification occurs when a layer of a less dense LNG overlies
a layer of denser LNG.
o
Stratification may develop:
•
When a less dense LNG is introduced on top of existing
tank contents that have a greater density.
•
When a denser LNG is introduced beneath existing tank
contents that have a lighter density.
•
When normal convection is interrupted by nitrogen
vaporising from the circulating liquid to create a surface
layer that is less dense than the rest of the tank
contents. This only likely to occur in LNG with a high N2
content. o
Unstable stratification occurs when the upper, less dense
layer, becomes gradually denser due to evaporation, and the
lower, denser layer warms and becomes less dense.
Density-stratified Liquid Layers
R
The addition of a different
batch of LNG can result in the
formation of strata of slightly
different temperature and
density within the tank.
R
Natural convection will occur in each layer individually. in each layer individually.
R
The upper liquid layer
evaporates as usual, but heat
in lower layer is trapped
beneath the upper layer and
cannot be released. As a
result, the lower layer will
become superheated & lighter
than upper layer.
Rollover
R
Superheating of lower layer results in
equilibrium of densities of the two
layers. When this occurs the interface
between the layers breaks down
resulting in a rapid transfer of heat and
mass within the storage tank.
R
The two layers mix rapidly and the
lower layer, which has been
superheated, gives off large amounts of
vapor as it rises to the surface of the
tank.
R
This phenomenon is known as “ROLLOVER”.
R
The large amounts of vapor generated
by this phenomenon can cause a
dramatic vapor expansion and increase
in internal tank pressure.
Industry Experience of Rollovers
R
There have been a significant number of rollover incidents in
shore LNG tanks.
R
Between1970 & 1982 GIIGNL received reports of 41 rollover
incident in shore tanks, the majority involved loading a cargo of
one composition into a tank containing heel of a different
composition.
R
Rollover is now widely understood in shore plants and
stratification is avoided by mixing different density liquids using
jet nozzles, recirculation, distributed fill systems, and alternate
top and bottom filling.
R
There have been fewer recorded incidents of rollover on board
LNG ships. Ship movement, the shape of the ship’s tanks and the
infrequent loading of LNGs of different density into the same tank
are factors that mitigate against the stratification of LNG in ships
tanks.
ROLLOVER Incident La Spezia
R
One of the more notably rollover incidents occurred at the LNG
receiving terminal in La Spezia, Italy, in 1971.
R
The actual pressures realized were unknown since they exceeded
available instrumentation and gas vented uncontrolled in a highly
populated area for several hours. Rollover occurred 31 hours
after loading from the Esso Brega.
R
Fortunately, there was no explosion and no major damage was
R
Fortunately, there was no explosion and no major damage was done to the tanks.
S.S. DEWA MARU
125,000m3 Moss-type LNG Carrier
Date of Launched: July 8th 1983
Voyage:
From: SAKAI, Japan (May 28th 2008)
To: PRIGORODNOYE, Russia (June 04th 2008)
DEWA MARU Voyage Background
R
THE PREVIOUS VOYAGE:
- Loading Port: Point Fortin, Trinidad & Tobago (06-7/04/2008).
- Discharging Port: Gwangyang, Korea (10-11/05/2008).
-
8,465m3 of LNG was retained onboard as Heel in No.3 & No.4 tanks
for a 48 day voyage toward next loading port, Damietta (Egypt). for a 48 day voyage toward next loading port, Damietta (Egypt).
R
CHANGE OF VOYAGE ORDERS
- The vessel was assigned for trade from Sakai, Japan to
Prigorodnoe, Russia on May 19th, 2008.
- The vessel arrived at Sakai LNG terminal on May 26th, 2008.
- Total 5,391m3 of LNG remaining onboard as Heel in No.3 &
No.4 tanks.
Voyage Synopsis
Event Date Comment
Depart Load Port , Point
Fortin, Trinidad
07 April 2008
Arrive Discharge Port
Gwangyang, Korea
10 may 2008
Depart Discharge Port
11 may 2008
8465m
3
heel for 48
Depart Discharge Port Gwangyang, Korea
11 may 2008
8465m
3
heel for 48
day ballast passage
Change of Orders 19 May 2008
Arrive Load Port Sakai,
Japan
26 May 2008
Depart Load Port Sakai,
Japan
28 May 2008 5391m
3
heel
Arrive Discharge Port
Prigorodnoye, Russia
02 June 2008
Berthed Prigorodnoye,
Russia
04 June 2008
Loading at Sakai
R
Sakai is principally a receiving terminal and not equipped to load
LNG ships at the rates normally expected of loading terminals.
R
The loading rate was max. 4,200m3/h instead of 11,000m3/h
due to the terminal loading pump capacity.
R
The equator temperature must be less than
-
134degC when the
liquid level reaches 1 metre below the equator level. The vessel The equator temperature must be less than
-
134degC when the
liquid level reaches 1 metre below the equator level. The vessel would normally conducts spraying operations during loading in
order to cool the tank to meet this requirement.
R
In this instance the terminal vapor system could not accept mist
in the vapour returned from the vessel and therefore the vessel
was not able to conduct spraying operations during the loading.
R
The vessel stopped loading each tank for 2-7 hours at 1 metre
below the equator level to ensure that the equator was cooled to
below -134degC.
R
The above factors may have contributed to the formation of
Density- stratified Liquid Layers in the heel tanks.
Closing CTM at Sakai
R
On completion of loading there was a big differential between T2
(85% tank height) and T3 & T4 (50% height and tank bottom) in
No.3 & No.4 Tanks
Voyage from Sakai to Prigorodnoye
R
Wind Beaufort 5 –7
R
Swell 1 –2 metres
R
Very little rolling or vibration
R
Low Voyage Speed 11.8 knots
R
Increase in liquid levels noted in
tanks 3 & 4
Tank Level Trend
1. Liquid level in No. 3 increased until Noon on 29th.
2. Liquid level in No. 4 increased until Noon on 31st.
3. Once Rollover occurred the liquid level went down rapidly.
4. Liquid level became steady when the circumstances in the tank had stabilised.
ROOOLOOOOOO ROOOLOOOOOO ROOOLOOOOOO ROOOLOOOOOO
Temperature and Pressure Sensors
No.3 Tank Liquid/Vapor
Temperature Trend
1.
Vapor Temp cooled since the liquid level came down in Tank 3 from Noon
on 29th, which indicates the generation of significant boil-off gas.
2.
AM on 02nd June, Rollover occurred in No.3 tank, the graph below
indicates that the upper / lower liquid temp. rapidly changed and also
that the vapor temp rapidly cooled down again.
No.4 Tank Liquid/Vapor
Temperature Trend
R
1. PM on 03rd June, Rollover occurred in No.4 tank, the below graph
indicates that the upper / lower liquid temperatures changed rapidly and
also the vapor temp rapidly cooled down again.
Tank Pressure Trend
1. Tank No.3 pressure trend showing pressure rise from rollover in
Tank No.3 on June 2
nd
and subsequent rollover Tank No.4
Opening CTM at Prigorodnoye
Proactive Measures to maintain safe tank pressure
R
When the crew detected that Rollover had occurred
in No.3 tank and No.4 tank, all vapor suction valves
except No.3 & No.4 tanks were closed up to min. %
necessary to maintain safe tank pressures in 1, 2 &
5 tanks to remove as much boil off gas as possible
from the “rollover” tanks.
R
On berthing the liquid temperature & tank pressures
were not yet safe level to conduct the opening CTM
and/or ESD Trip test. The master of the vessel
coordinated with terminal to send BOG to shore flare
to reach stable condition as soon as possible.
Events at Prigorodnoye
P
08:35 2
nd
June: Anchored awaiting clearance
P
09:00 2
nd
June: Rapid rise in Tank 3 pressure to 17.5 kpa –
rollover – full steam dumping
P
13:50 3
rd
June: Rapid rise in Tank 4 pressure to 18.5 kpa then
drop off
P
14:00 3
rd
June: Weigh anchor to proceed to LNG anchorage
P
14:00 3
June: Weigh anchor to proceed to LNG anchorage
P
14:30 3
rd
June: Start spray pump in Tank No.4 to encourage
intentional rollover under controlled conditions – stopped when
tank pressures reach 18 kpa
P
15:40 3
rd
June: Anchored at LNG Anchorage – No.4 tank
pressure rose to 20 kpa – rollover occurred.
P
00:00 4
th
June: Tank pressures stabilised and start to fall.
Master agrees with terminal to send vapour to flare until tank
pressures / temperatures stabilise sufficiently to allow CTM.
P
10:00 4
th
June: All fast
P
13:20 4
th
June: Commence sending vapour to shore flare
P
14:48 4
th
June: Stop sending vapour to flare
P
15:08 4
th
June: Opening CTM
Lessons Learnt
R
It was believed that Rollover would be unlikely to occur in a Moss
type tank because the spherical shape of the tank would aid the
migration of the warmed liquid along the tank wall.
Lessons Learnt
Compared with events at La Spezia the consequence of the
stratification experienced by the Dewa Maru was more a ‘fold-over’
than a rollover. The tank pressures were maintained within the design
parameters at all times however the incident demonstrated that:
R
Stratification and rollover conditions can develop on board ship
particularly when loading a higher density LNG into a tank containing
a lighter cargo.
R
Ship movement on passage cannot be relied upon to mix layers of
dissimilar density.
R
Increase in tank levels may be an indication of stratification. Spherical
tanks act as a ‘lens’ for readily noting changes in volume.
R
Reduction in boil off gas generation may indicate stratification.
Change in Tank Level for Change
in Volume
C
40 3
11 3
40 3
Thank you for the opportunity to
make this presentation:
Now may we take your questions?