HIBERNIA MANAGEMENT AND DEVELOPMENT COMPANY LTD FLARE SYSTEM REVALIDATION STUDY TECHNICAL NOTE DOCUMENT NO : 8266-HIB-TN-C-0001 REVISION : B DATE : October 2000 8266-HIB-TN-C-0001 Page of 204 Revision: B /home/website/convert/temp/convert_html/55276c7749795994178b46d5/document.doc October 2000 1
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
HIBERNIA MANAGEMENT AND DEVELOPMENT COMPANY LTD
FLARE SYSTEM REVALIDATION STUDY
TECHNICAL NOTE
DOCUMENT NO : 8266-HIB-TN-C-0001
REVISION : B
DATE : October 2000
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
1
DOCUMENT REVISION RECORD
REV DATE DESCRIPTION PREPARED CHECKED APPROVED
Draft 21/07/00 Issued for IDC M. Goodman A. Robinson M. Goodman
A 23/08/00 Issued for Comment / Final M. Goodman A. Robinson M. Goodman
B 16/10/00 Final M. Goodman A. Robinson M. Goodman
RELIANCE NOTICE
This report is issued pursuant to an Agreement between Granherne (Holdings) Limited and/or its subsidiary or affiliate companies (“Granherne”) and HIBERNIA MANAGEMENT AND DEVELOPMENT COMPANY LTD which agreement sets forth the entire rights, obligations and liabilities of those parties with respect to the content and use of the report.
Reliance by any other party on the contents of the report shall be at its own risk. Granherne makes no warranty or representation, expressed or implied, to any other party with respect to the accuracy, completeness, or usefulness of the information contained in this report and assumes no liabilities with respect to any other party’s use of or damages resulting from such use of any information, conclusions or recommendations disclosed in this report.
Title:
FLARE SYSTEM REVALIDATION STUDY
QA Verified:
TECHNICAL NOTE Date:
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
2
CONTENTS
FRONT PAGE
DOCUMENT REVISION RECORD
CONTENTS
ABBREVIATIONS
HOLDS
1.0 INTRODUCTION
2.0 SUMMARY AND CONCLUSIONS
2.1 Introduction
2.2 Technical Audit of the Design Calculations
2.3 Challenge Process
2.4 As-Building the Flare System
2.5 Risk Management in Relation to Wind Condition and Flaring Events
2.6 Implications for Hibernia
2.6.1 Introduction
2.6.2 Capacity Opportunities
2.6.3 Impact on the Design Documentation
2.6.4 Future Work
3.0 DESIGN BASIS
3.1 Introduction
3.2 Safety Design Basis
3.2.1 Probabilistic Design Criteria
3.2.2 Deterministic Design Criteria
4.0 APPROACH
4.1 General
4.1.1 Flare System Revalidation Process
4.1.2 Legislative Obligations of HMDC’s Safety Design Philosophy
4.1.3 The Requirements of the Standards, Codes of Practice and Recommended
Practices
4.1.4 Ambiguities in the Recommended Practices
4.2 Calculation Audit
4.3 Challenge Process
4.4 Risk Management in Relation to Flaring Events and Wind Condition
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
3
5.0 TECHNICAL AUDIT OF THE DESIGN CALCULATIONS
5.1 Introduction
5.2 Results of the Technical Audit – Relief and Blowdown System Calculations
5.2.1 Technical Audit Issue Discussion – Relief and Blowdown System Calculations
5.3 Results of the Technical Audit – Relief Valve Sizing Calculations
Reliability analysis of the system that controls the compressor stagger, to ensure
the system is sufficiently reliable to ensure the design integrity.
Implementation Projects
In this section there are some projects mentioned which will in all likelihood require
hardware changes to be made (resulting from the above there may be more).
Insulation conformance - The explicit ability of the platform to cope with a jet fire
hazard requires the insulation around the vessels to remain in place during jet
flame impingement. This may require the insulation strength to be improved.
Modifications to limit peak flaring rate during spillover valve failure.
Instrument modifications to warn operators when the requirement to blowdown
compressors is becoming imminent (to avoid low temperature problems).
In discussion with HMDC a detailed scope of work to undertake the above has been
developed. This is attached in Appendix 2.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
17
3.0 DESIGN BASIS
3.1 Introduction
The review focuses on those aspects and hazards that are directly relevant to the flare
system design. In particular these are flare radiation, jet and pool fire and, to a lesser
extent, explosion. This naturally excludes issues relating to blowout and iceberg
collisions as well as environmental issues.
However, before going into the main parts of the analysis, it is worth recapping the
safety basis and methodology followed by HMDC now and during the design phase.
This will need to be followed should any changes be made to design philosophies.
3.2 Safety Design Basis
As was convention at the time, the safety design progressed along two parallel routes.
The first was the use of probabilistic analysis to identify the acceptability of various
risks. The second was the deterministic design of the various safety systems
according to recognised codes and practices. Occasionally there was an interface
between the two processes when a risk was considered unacceptable. Where this
was apparent the design would be adjusted to mitigate the unacceptable risk.
These two processes are described in sections 3.2.1 and 3.2.2 below.
The problem of parallel processes is that some information can be lost across the
interface. More recently this has led to a concept called risk based design where the
key safety issues are resolved during the early conceptual design stages rather than
be left for implementation after the conceptual design is complete.
3.2.1 Probabilistic Design Criteria
By the time the FRA was commenced the HMDC Damage / Impairment Criteria had
been formalised. These were:
Criterion 1: Overall Platform Integrity
There must be no overall loss of integrity of the platform for at least 2 hours after the
initial event. Loss of integrity included:
Structural collapse of more than 50% of the platform topsides, or total collapse of
Module M30.
The 2 hours is judgementally used for a maximum time to evacuate by lifeboats (i.e.
time to respond to emergency, attempt to control, organise evacuation and abandon
platform)
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
18
Criterion 2: Integrity of Temporary Safe Refuge (TSR)
The TSR (i.e. living quarters) should remain a safe refuge for personnel for at least 2
hours. Loss of integrity may be due to:
Fire within the safe refuge;
Blast damage, in excess of major window breakage;
Collapse of any part of shelter area.
The time of 2 hours is derived and defined as for Criterion 1.
Criterion 3: Escape Routes
Escape routes from all parts of the platform to the TSR or other safe refuge should
remain passable for at least 30 minutes from the start of the incident. An escape route
may be made impassable by:
Thermal radiation over 12.5 kW/m2 to the outside of the escape route if protected by cladding;
Thermal radiation over 6.3 kW/m2 if unprotected;
Blockage due to blast damage;
Collapse of one or more modules;
Flooding over 1 m deep in the Utility Shaft.
The time of 30 minutes is intended to allow the escape of workers who had initially remained at their posts to shut-down the process operation or fight a growing fire. The criterion is violated if an incident results in either:
All escape routes from any module being impassable;
All routes from any one part of the platform to the TSR being impassable.
Since there is more than one escape route from any point, an incident must completely involve the total module to violate the criterion.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
19
Criterion 4: Means of Evacuation
The evacuation systems must remain effective for long enough to evacuate all
personnel. This requires at least one of the following to be true:
Helideck operable for at least 2 hours. Inoperability may result from one of the
following:
tilt over 15°
smoke due to oil fire and wind towards helideck
thermal radiation over 3.2 kW/m2
blast damage
unignited gas over helideck (due to likelihood of ignition)
collapse of M50 module.
Evacuation systems must be operable with at least 10% spare capacity (to allow for
launching partly loaded) and with passable escape routes, from TSR to evacuation
system, for at least 2 hours. Inoperability may result from:
tilt over 25° (preventing safe access)
thermal radiation over 12.5 kW/m2
blast damage (damaging launching gear and access walkways)
collapse of module M40 and M30.
The times are judgementally based on the proposed systems for the Hibernia platform.
The criterion is only violated if all the means of evacuation are unavailable.
In general the following approach was applied:
The Damage / Impairment Criteria set out above, give basic criteria which should not
be exceeded. It is not possible to ensure that no incidents will exceed the criteria.
The intent is that every reasonable and practical precaution is taken to ensure those
incidents that exceed the criteria are so unlikely that they can be considered as an
acceptable risk because the risk is negligible. These incidents are termed Residual
Accidental Events.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
20
Added to this were Hibernia’s three-tier framework of risk acceptability:
For any single incident that might affect the key safety systems (more accurately
safety functions from the above), the risk level for the three-tiers are:
Intolerable: greater than 10-4 per year.
ALARP region: 10-4 to 10-5 per year.
Lower bound of acceptability: less than 10-5 per year
Whilst it is inconceivable that any of the impairment criteria would change as a result
of the considerations in this report, change may affect the QRA upon which these
impairment criteria stand. Any changes considered, therefore, will require to be
confirmed through QRA.
3.2.2 Deterministic Design Criteria
A number of the final requirements for the design would stem from the above. This is
not surprising as some of the aspects of the impairment criteria actually have their
roots in the recognised international codes and practices, e.g.
Allowable Flare Radiation Levels:
Escape Ways
Not over 12.5 kW/m2 to the outside of the escape route if protected by cladding;
Not over 6.3 kW/m2 if escape way is unprotected
Helideck
Not over 3.2 kW/m2
The remaining requirements are part of the various design guides and codes of
practice. These are described in detail in Section 6.0
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
21
4.0 APPROACH
4.1 General
4.1.1 Flare System Revalidation Process
The flare revalidation process is summarised in the flowscheme overleaf. The stage
consisting of this report is Stage 1. The results of Stage 1 will form the basis for future
Stage 2 studies. The flowscheme indicates the potential range of projects this could
encompass.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
22
Figure 4.1 - Revalidation Process
Start
Technical audit of the existing
design calculations Challenge process Risk management
Prepare Flare System Revalidation
Report
Draft and Final Versions
Impact on
the design?
Update Relief and
Blowdown Study Report
(Rename “Flare System
Design Philosophy”)
Build flare network model
(for inclusion in the Flare
System Design
Philosophy) (Optional)
Stage 1
Major changes to flare
design calculations. As
build and replace the
design calculation volumes
Update Relief and
Blowdown Study Report.
(Rename “Flare System
Design Philosophy”)
Build flare network model
(for inclusion in the Flare
System Design Philosophy)
Prepare workscopes for the
modifications
Minor changes to flare
design calculations. Rev up
affected calculation
volumes
Update Relief and
Blowdown Study Report.
(Rename “Flare System
Design Philosophy”)
Build flare network model
(for inclusion in the Flare
System Design Philosophy)
(Optional)
End
Major impact Minor impact
Stage 2
Varies according
to outcome of
Stage 1
No Impact
HMDC Internal Audit /
Approval
HMDC Internal Audit /
Approval
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
23
4.1.2 Legislative Obligations of HMDC’s Safety Design Philosophy
HMDC’s legislative obligations to safety are encompassed in the Newfoundland
Offshore Petroleum Installations Regulations (Reference 1), an extract of which
follows:
43. (1) Every operator shall…submit to the Chief a concept safety analysis…that
considers all components and all activities associated with each phase in the life of the
production installation, including the construction, installation, operation and removal
phases…
(5)…
…(g) a definition of the situations and conditions and of the changes that would
necessitate an update of the concept safety analysis.
(8) The operator shall maintain and update the concept safety analysis referred to in
subsection (1) in accordance with the definition of situations, conditions and changes
referred to in paragraph (5)(g) to reflect operational experience, changes in activity or
advances in technology.
HMDC have met these requirements, initially through the preparation of the Concept
Safety Evaluation which has, over time, evolved into the Operational Plan which will be
issued in the near future.
The above very much parallels the type of approach the UK HSE require:
11. The employer…needs to review the risk assessment if there are developments
that suggest that it may no longer be valid (or that it can be improved). In most cases,
it is prudent to plan to review the risk assessments at regular intervals - the time
between reviews being dependent on the nature of the risks and the degree of change
likely in the work activity. Such reviews should form part of standard management
practice.
Management of health and safety at work - Approved Code of Practice (1992)
The study will be undertaken in this light.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
24
4.1.3 The Requirements of the Standards, Codes of Practice and Recommended
Practices
Where a particular design code is used its requirements are mandatory, e.g. the
requirements of ASME VIII for a vessel stamped accordingly. Recommended
practices are different in that their requirements are not mandatory in law unless they
are stated in the regulations; this is the case with Canadian requirements. For those
practices not in the regulations, common industry practice and deviation from those
would normally require Certifying Authority approval. In dispute, the applicability of the
use of the practice would be left to the courts to decide.
So, although recommended practices are not the same as design codes, their
requirements have become almost code like over time. Consider a situation where a
failure has occurred and its cause appears to be linked to a situation where a well
known, recommended practice was deviated from. There would in effect be an onus
to prove the requirement inferred in the practice was inappropriate at the time it was
being considered. Otherwise negligence would be very difficult to disprove. This proof
would be particularly hard to provide and would require thorough documentation
regarding the deviation to be kept for the life of the plant. This study will recognise this
reality and, therefore, design code, code of practice and recommended practices, so
long as they emanate from a recognised responsible body, are considered equivalent
in this study in terms of reliance.
4.1.4 Ambiguities in the Recommended Practices
The recommended practices are sometimes (some would say often) ambiguous in
their requirements and a study such as this tends demonstrate the problem. Therefore
interpretations of the practice’s actual intent often have to be made. This is one of the
designer’s challenges and a particular challenge of this report.
In the past, presumably to avoid the need for interpretation, owners of facilities have
removed the ambiguities by being prescriptive with their requirements. Thus, industry
practices, which often have little connection with the original design code intent, have
sometimes been established for expediency. The goal today is to apply the codes as
intended without unnecessary features that increase cost.
The methodology used in this study therefore lies in this latter approach to the
application of practices, to apply them as intended. Where there are differences
between the various methods of applying practices this will be highlighted in the
narrative.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
25
4.2 Calculation Audit
The calculation technical audit’s primary aim is to identify key assumptions or queries
contained outside the Relief and Blowdown Study Report. A secondary aspect is the
high level numerical and methodological check of the existing relief case calculations
to ensure their suitability to use as the basis for the revalidated Flare Design
Philosophy. This process will identify any areas that require detailed review to be
undertaken in future stages of the revalidation study.
The audit will use a tabular approach (compiled by calculation volume) to highlight the
assumptions or issues which require to be addressed during the challenge and risk
mitigation review processes in subsequent sections of the study.
Whilst addressing the calculations, secondary objectives such as consistency and
methodology have also been revisited and these results can also be found on the
detailed audit sheets.
4.3 Challenge Process
The challenge process is a sequential review of individual design parameters that had
an effect on the way the flare system was dimensioned.
The process looks first at the requirements of the design codes or practices to
place, in an historical context, the requirements for the design. The key codes and
practices applicable to this work are those referred to in the RABS, i.e.:
Canada Oil and Gas Installation Regulations
Petroleum Occupational Safety and Health Regulations - Offshore
Newfoundland
API RP 520
API RP 521
API RP 14C
Mobil EGS 661
ASME Section VIII, Division 1
Secondly, how the system was actually designed is considered. This aspect
captures the interpretations used when compared to the earlier activity.
Thirdly, the requirements of current codes and practices are then considered.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
26
Fourthly, current best practice is considered to challenge the existing design and
determine its suitability for application to Hibernia. Here the best practice applied
is Granherne’s own (there is no other convincing way for us to address this issue).
This is not to say that other companies do not apply the requirements differently.
Where possible some alternative applications will be mentioned where relevant.
Finally, the effect of these stages is considered and a recommendation made for
the way the requirements should be applied to Hibernia to give a consistent and
easily understood flare system design.
4.4 Risk Management in Relation to Flaring Events and Wind Condition
This issue stems from the technique used occasionally where the capacity of a flare
system has been increased when it has been realised that at the design windspeed no
personnel would be present on deck, thereby allowing higher incident radiation rates
on deck during these events. This would only be the case if a very high wind speed
were considered during design. A more pragmatic approach to design windspeed
selection would ensure that the coincident conditions were considered, i.e. a
realistically high windspeed.
Generally two cases are considered:
Emergency flaring
The methodology used to generate the results of the activity are based on multipliers
applied to the flowrates considered in the original flare boom length defining design
case, i.e. combined HP and LP blowdown. It should be appreciated that these
resultant rates cannot actually occur until the platform is actually modified to
incorporate the necessary inventory, in this case, in the same ratio (HP/LP) as design.
This is unlikely. The real effects, or real envelope, will therefore depend on the actual
modification made. The actual effect should be considered in detail during the
particular modification project’s design phase.
Continuous flaring
Continuous flaring differs from emergency flaring in that the acceptable radiation levels
are very much lower than for emergency events. Proportionally, the lower radiation
isopleths are more sensitive to wind than the high flow cases.
The methodology used in this study has taken a simulation from the recent
debottlenecking project (Case 3, a case which included Avalon production) and used
this to generate current normal operating input data to generate a new set of profiles.
The input data is adjusted up or down using simple multipliers to generate the
expected envelope.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
27
5.0 TECHNICAL AUDIT OF THE DESIGN CALCULATIONS
5.1 Introduction
The intent of this review is twofold:
To identify assumptions or items which have affected the design of the relief and
blowdown system including any issues which arise as a result of the review
(Section 5.2).
To revisit the relief valve calculations to perform a methodological and numerical
check to reconfirm the validity of the dimensioning design cases (Section 5.3).
5.2 Results of the Technical Audit – Relief and Blowdown System Calculations
In Appendix I the full results of the audit are given. The detailed tables that follow
identify a number of issues to be dealt with, either in Section 6.0, because they can be
challenged, or in this section if they are issues which concern the accuracy or
soundness of the design conclusions. For completeness, however, both sets of issues
are summarised below.
Table 5.3 Challenge Issues - Relief and Blowdown System Calculations
(System 34)
Calculation Issue
Number
34-
Title Rev Date Number
34-
Description
005 / A Blowdown Section Inventory Calc (Provides input to blowdown simulations)
06 18-May-93 005/2 Jet fire scenario not taken into account for
C1 Nov-91 31.35/1 Does 2 phase relief case become the governing case if the calculation new calculation method given in API RP520, Seventh Edition used?
31.35/2 Flare network analysis for 2 phase case
(Calc 34-064 / G) used total load = 252,372
kg/h (40,000 bpd).
31.35/3 Relief & Blowdown Study Report Rev C1
states HP Separator Blocked Outlet
(Vapour) relief load is 244,897 kg/h.
31.35/4 The two phase calculation feed vapour /
liquid split was abnormally low.
31.35/5 Methodological problem in calculation
(compared to API RP520 Sixth Edition).
The wrong effective pressure was for the
V/L split and property conditions.
31.36 Relief Valve Calculations - MP Separator
C1 Nov-91 31.36/1 Does 2 phase relief case become the
governing case if the calculation new
calculation method given in API RP520,
Seventh Edition used?
31.36/2 Are 2 x 50% LCVs sufficiently independent?
31.36/3 Methodological problem in calculation
(compared to API RP520 Sixth Edition).
The wrong pressure was used to generate
the vapour amount and properties.
31.36/4 The two phase calculation feed vapour /
liquid split was abnormally low.
31.36/5 Calculation subsequently superseded but
no indication that calculation was
subsequently corrected.
31.36/6 The gas blowby cases are methodologically
flawed.
31.36/7 There is an error in the gas rate calculated
by the test separator gas blowby case.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
1580 For continuous flaring operations in areas where personnel must remain at
their work stations without shielding but with appropriate clothing
1580 Emergency flaring for several minutes(2) - personnel without appropriate
clothing
790 Continuous flaring(2) - personnel expected to wear appropriate clothing
3155 Emergency flaring up to one minute(2) - personnel without appropriate
clothing
Notes
1. In areas where personnel can be exposed to higher radiation intensities, heat shielding must be
provided and also for equipment and structure as necessary.
2. In areas where personnel are not expected to wear appropriate clothing (i.e. coveralls, boots,
gloves, hard hats) allowable radiation levels have been reduced by a factor of two.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
104
API RP 521 (Third Edition, November 1990) recommends the following:
Table 6.21 Recommended Design Flare Radiation Levels Excluding Solar
Radiation (API RP 521)
Permissible Design
Level (K)
KW/m2
Location
15.77 Heat intensity on structures and in areas where operators are not likely to be
performing duties and where shelter from radiant heat is available (for example,
behind equipment).
9.46 Value of K at design flare release to any location to which people have access (for
example, at grade below the flare or a service platform of a nearby tower);
exposure should be limited to a few seconds, sufficient for escape only.
6.31 Heat intensity in areas where emergency actions lasting up to 1 minute may be
required by personnel without shielding but with appropriate clothing.
4.73 Heat intensity in areas where emergency actions lasting several minutes may be
required by personnel without shielding but with appropriate clothing.
1.58 Value of K at design flare release to any location where personnel are continuously
exposed.
6.7.2 The Radiation Levels Used in the Design
The design used the following radiation levels, derived from Draft Canadian
Legislation, as outlined by the RABS:
Table 6.22 Radiation Flux Limits Excluding Solar Radiation
Permissible Design
Level (K)
KW/m2
Conditions
6.3 Heat intensity in areas where emergency actions lasting up to 1 minute may be
required by personnel without shielding but with appropriate clothing.
4.72 Heat intensity in areas where emergency actions lasting up to several minutes may
be required by personnel without shielding but with appropriate clothing.
1.9 Value of allowable radiation level at design flare release at any location where
personnel are continuously exposed, i.e. helideck
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
105
In addition the following statement is included in the RABS:
In addition to the above radiation limitations HTPT advised that the maximum radiation
level experienced on the platform escape routes is not to exceed 1000 Btu/ft2 h (3.16
W/m2) for periods over 1 minute of exposure.
RABS page 22
The design calculations for the worst emergency flaring case (total platform blowdown)
and a flare boom length of 115m resulted in the following radiation levels:
Approximately 3.16 kW/m2 at the north side M10 weather deck
Approximately 6.30 kW/m2 at the drilling derrick crown block
Approximately 4.72 kW/m2 at the drilling derrick finger board
6.7.3 Current Requirements of the Codes and Recommended Practices
6.7.3.1 Mobil “Pressure Relief and Vapor Depressuring Systems” MP 70-P-06, July 1998
The new document has the following recommendations on thermal radiation levels.
Table 6.23 Allowable Radiant Heat Intensities in W/m2 Excluding Solar Radiation
Appropriate Clothing* Without Appropriate Clothing*
Continuous Release 1105 790
Emergency Releases
Travel time to Shelter
6 Sec 9465 3150
1 Min 6300 3150
3 Min 4730 1580
No Shelter Available 1580 790
Equipment Only
Exposure15770
Volatile Liquids Tanks,
API Separators, CCB
2365
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
106
6.7.3.2 API 521 (Fourth Edition, March 1997)
The recommendations on flare thermal radiation levels in the new addition of API
RP521 remain the same as the previous version.
6.7.4 Current Best Practice
We have mentioned earlier in this report that best practice is subjective to some
extent. Where issues are not subjective are in matters of law. Once a requirement
passes into law, as have the Canadian regulations, by meeting those requirements, an
owner has effectively discharged their responsibilities. As is also customary in matters
of precedence, national regulations always supercede recommended practices.
Normally there is actually little difference between the two requirements and this is
where we find ourselves in the Hibernia context. This is demonstrated in the following
table:
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
107
Table 6.24 Summary of Flare Radiation Requirements for Hibernia
Canadian Regulations
Impairment Criteria
Equivalent API RP521
Remarks
Maximum radiation on areas where the period of exposure will be greater than one hour
1.9 N/A 1.58 Apply Canadian regulations.
Helideck operable for at least 2 hours. Inoperability may result from…thermal radiation over 3.2 kW/m2
(Impairment Criterion 4)
Silent 3.2 Not specifically mentioned
Not used as a normal radiation level.
Maximum radiation on areas where the period of exposure will be greater than one minute but not greater than one hour
4.72 N/A 4.73 Apply Canadian regulations.
Maximum radiation on areas where the period of exposure will not be greater than one minute
And,
Escape routes from all parts of the platform to the TSR… to remain passable for 30 minutes…An escape route may be made impassable by:
Thermal radiation over 6.3 kW/m2
if unprotected:
(Impairment Criterion 3)
6.3 6.3 6.3 Apply Canadian regulations.
Maximum radiation on areas where the period of exposure will not be greater than a few seconds
(In this case the area in question is normally accessible)
Silent N/A 9.5 Apply API requirements in absence of Canadian regulation.
The actual wording of API 521 is:
Value of K at design flare release to any location to which people have access (for example, at grade below the flare or a service platform of a nearby tower); exposure should be limited to a few seconds, sufficient for escape only. (Note 1)
Escape routes from all parts of the platform to the TSR… to remain passable for 30 minutes…An escape route may be made impassable by:
Thermal radiation over 12.5 kW/m2 to the outside of the escape route if protected by cladding:
(Impairment Criterion 3)
Silent 12.5 Not specifically mentioned
Not used as a normal radiation level.
Maximum radiation on areas where shelter is present.
(In this case the area in question is not normally accessible)
Silent N/A 15.8 Not used as a normal radiation level.
The actual wording of API 521 is:
Heat intensity on structures and in areas where operators are not likely to be performing duties and where shelter from radiant heat is available (for example, behind equipment). (Note 1)
Notes:
1_) On towers and elevated structures where rapid escape is not possible, ladders must be provided on
the side away from the flare, so the structure can provide some shielding when K is greater than …
6.3 kilowatts per square meter.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
108
The allowable radiation levels on Hibernia will have to be selected from within these
requirements.
6.7.5 The Effect of Applying Best Industry Practice to Hibernia
The application of Canadian regulations and interpretation of the guidelines in API
RP521 where the Canadian requirements are silent (replicated in Table 6.24) for the
Hibernia platform gives the following allowable thermal radiation levels at various parts
of the platform:
Crown Block 9460 W/m2
The crown block falls into the category an area to personnel have access (i.e. a
service platform of a nearby tower); where exposure can be limited to a few seconds,
sufficient for escape only.
It could even be argued that a more extreme limit at this point could be used: The
Damage / Impairment criterion No. 3 indicates that a value of 12500 W/m2 may be
appropriate for the area under consideration. The criterion is specifically aimed at
escape routes protected by cladding but could equally be applied to the drilling derrick
which is partially enclosed and offers any operator working in the area the opportunity
to shelter behind a clad structure for the duration of the emergency. A reference for
the figure of 12500 W/m2 cannot be found in the guides and practices referenced
above although a somewhat worse value of 15800 W/m2 can be found in the API
which is allowed only in an area where shielding exists. These requirements are
included for information only.
In the original design it appears an unnecessarily conservative approach, which did
not recognise the presence of shielding, was applied to this area which limited the
radiation level to 6300 kW/m2.
Weather Deck 6300 W/m2
The weather deck and monkey board falls into the category of an area where
emergency actions lasting up to one minute may be required by personnel without
shielding but with appropriate clothing. It is expected that personnel on the weather
deck would be appropriately clothed and in the event of an emergency blowdown
would be able to leave either leave the deck in a minute or less or alternatively find
shelter in the same time period.
In the original design, values of 3200 and 4720 kW/m2 respectively were applied to
these areas. The former resulted from the HTPT note attached to the table which
outlined the explicit design requirements and, from the above, is a radiation level
allowable for 2 hours in an emergency. The latter neither recognises the escape
ability from the monkey board nor the shielding. Both cases therefore appear
unnecessarily conservative.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
109
Helideck 3200 W/m2
The helideck does not really fall into any specific category as defined in the guides and
practices reference above but it could be argued that it loosely falls into the category of
an area where personnel are continuously exposed during an emergency (for up to
two hours) and therefore the value of 3200 W/m2 is chosen. This corresponds with the
Damage / Impairment criterion No. 4 which indicates that a value of 3200 W/m2 is
appropriate for the helideck which is based on Canadian regulations.
Weather Deck (Continuous flaring) 1900 W/m2
For a true (non-emergency) continuous release, e.g. production flaring, a figure of
1900 kW/m2 should be used (again in accordance with Canadian regulations).
Of the above, the most important radiation level is likely to be the continuous flaring
case as it will be the most persistent (occasionally). The other radiation levels are only
approached during a platform blowdown and therefore are short duration (only
seconds) and will only be felt if a coincident severe adverse wind occurs during the
event.
Using this radiation level and location as the design case ensures that the helideck will
experience very much lower radiant rates during continuous flaring.
The radiation levels used in the flare operating envelope calculations, described fully in
Section 8.0 below, have used the above thermal radiation limits to determine the
allowable maximum flaring capacity for the ‘As Built’ flare for two windspeeds.
The results of these calculations are discussed in detail in Section 8.0 but the principle
conclusion is that if the best practice radiation levels are applied as defined above then
the flare system capacity would be approximately 200% of the current design load in
terms of thermal radiation only. The effect on hydraulics in the system for this capacity
increase has not been studied at this time.
This section should be read in conjunction with Section 8.4.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
110
6.8 Challenge Issues Resulting from the Technical Audit of the Design Calculations
Issue 34-005/2 - Jet fire scenario was not taken into account for in the design of
the blowdown system.
This issue is addressed in Section 6.2.
Issue 34-005/4 - Were fire areas used for total blowdown rate?
A simple approach to blowdown was used where the entire all equipment to be
depressured was assumed to be on fire. This is equivalent to the entire M10 module
being on fire, something which is very unlikely and if it occurs will be catastrophic.
More conventionally the platform is separated into fire areas. In this case the
blowdown valves are sized to cater for the fire case. However, the combined case is
not normally the sum of all the areas on fire and some effort is instead focussed at the
selection of a realistic worst case. The worst case is represented by the fire occurring
in the area which adds most to platform load coincident with the resultant rates from
the blowdown valves for the non-fire areas are added. These latter rates are less than
the rate that would be experienced in a fire case and the overall blowdown load is
more accurately represented. In this case we have been unable to locate fire area
drawings which forces the M10 fire case to remain the design case.
Issue 34-006/2 - Is correct isentropic efficiency used?
The isentropic efficiency specified when performing blowdown simulations affects both
the downstream blowdown temperature of gas and equipment but also the upstream
vessel wall temperature. An isentropic efficiency of 1 simulates perfectly isentropic
expansion of the gas and gives the worst case (i.e. lowest) temperatures. An
isentropic efficiency of 0 simulates perfectly isenthalpic expansion of the gas and gives
the best case (i.e. highest) temperatures. For blowdown of a vessel or system where
the feed to the vessel has been stopped the expansion of the gas is somewhere
between isentropic and isenthalpic. The selection of the isentropic efficiency is usually
based on project philosophy and experience.
The blowdown simulations performed for these calculations used an efficiency of 0.5.
There is no indication in the calculations or simulation outputs for the basis of this
selection. The selection of an efficiency of 1 is unrealistic, but a more usual figure to
use is 0.7 minimum which would lead to lower blowdown temperatures.
The impact of lower blowdown temperatures is twofold. The first concerns the
materials of construction of the flare system itself. The flare system appears from the
‘As Built’ P&IDs to constructed of LTCS with a minimum design temperature of -45oC.
The second impact is in areas of the process where hydrates can form.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
111
From the calculations reviewed problems of both hydrate formation and flare design
temperature only occur if blowdown is initiated after a delay with of the plant
maintained at pressure during the upset. Calculation 34-010 / A and 34-060 / B
address this problem but the calculations do not give any specific conclusions on the
allowable delay. Calculation 060 / B concludes that for the settleout pressures used
there is a huge spread of allowable delay periods depending upon environmental
conditions and whether insulation is installed. Current platform design philosophy is to
depressurise after 1-2 hours. If lower blowdown temperatures are expected then this
philosophy may have to be reviewed. See Issue 34-010/1 for further details.
Precautions against hydrate formation can be taken and these are discussed further
below.
For more discussion regarding this see the related discussion in Section 5.2.1.2 item
34.010/1.
Issue 34-006/3 - Is design case too extreme?
This concern relates to the high start pressure used for the blowdown calculations.
See Section 6.3 for the recommended solution.
Issue 34-042/2 - Validity of staggering blowdown. Were the systems sufficiently
independent?
This aspect is covered in Section 6.4.
6.9 Miscellaneous Issues
6.9.1 Insulation
The presence of satisfactory insulation on the vessels will allow substantially reduced
depressuring rates compared to those used during design. This aspect should be
checked in detail when the insulation on the vessels is reviewed.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
112
7.0 AS-BUILDING THE FLARE SYSTEM
7.1 Introduction
The Hibernia platform was built incorporating features for future equipment; this
included future capacity built into the flare system. As some of these projects are no
longer foreseen this section looks to remove their effect from the currently installed
flaring cases. This, in effect, will result in a system whose design cases are “as-built”.
The difference between the design capacity and the “as-built” capacity is the capacity
available for future projects, including those that were originally foreseen.
7.2 As-built and Design Capacity
The following table summarises the initial relieving capacity by area considered during
the design phase as well as the results of removing the requirements for future
equipment and potentially the 3 minute stagger on the injection compressor ‘A’
blowdown.
Table 7.25 As Built and Design Capacity
Case Scenario LP Flare Load
kg/h
HP Flare Load
kg/h
1 Design Blowdown Rate as per Relief &
Blowdown Study Report Rev C1
89,601 133,616
2 ‘As Built’ i.e. As Case 1 with Future Equipment
Removed
89,601 94,843
3 As Case 2 with 3 min Stagger Removed 126,291 94,843
The table below summarises the effect on thermal radiation impingement at various
points on the platform for the cases described in Table 7.1 with the original design
wind speed of 27 m/s blowing in a northerly direction.
Table 7.2 Thermal Radiation Impingement on Platform Areas
Case Crown Block
W/m2
Weather Deck
W/m2
Helideck
W/m2
1 6152 2911 1034
2 5623 2700 860
3 8838 3146 1314
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
113
The capacity of the flare system is essentially decided by the allowable thermal
radiation impingement on the platform. The different levels of thermal radiation and
their limitations on working and escape routes are discussed in Section 6.7. Based on
the original design radiation levels:
Crown Block 6300 W/m2
Weather Deck 3200 W/m2
Helideck 1900 W/m2
Then the following deductions can be made from the resulting thermal radiation
impingement on the platform for the 3 flare operating cases described in Table 7.1.
An immediate and obvious conclusion which can be drawn from the results of this
study is that a wind speed of 27 m/s gives higher thermal radiation levels on the
platform than the 24.5 m/s wind speed for the relief cases considered here.
However if we analyse the results given for the wind speed of 24.5 m/s, the Granherne
best practice wind speed as defined in Section 6.6, figures 8.1 to 8.5 and 8.11 above,
the following conclusions can be drawn for each flaring scenario:
8.4.2 Emergency Relief - Platform Blowdown
Maximum allowable thermal radiation at the crown block - 9500 W/m2
For this case the thermal radiation on the crown block is the limiting factor. An initial
blowdown rate of approximately 150% of the design rate can be tolerated before the
limit of 9500 W/m2 is limit is exceeded in this area.
The other cases considered are less onerous, with the 3200 W/m2 isopleth impinging
on the Helideck at around 350% of the design blowdown rate and the 6300 W/m2
isopleth impinging on the weather deck at around 370% of the design blowdown rate.
The above suggests considerable capacity is inherent in the system dependent on the
final basis selected. However, caution should be exercised as this apparent capacity
will change dependent on the detail of the project which actually utilises the apparent
capacity. In other words the absolute capacity will only be confirmed once the LP and
HP rates are fully defined and detailed calculations are performed for the modification
under consideration.
8.4.3 Continuous Relief
For this case the maximum allowable thermal radiation of 1900 W/m2 at the weather
deck is considered to be the limiting factor. For a northerly wind blowing at 24.5 m/s, a
platform production rate of 62% of the design rate (considered to be 200 kbopd) can
be tolerated before the limit of 1900 W/m2 is limit is exceeded in this area.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
125
Here is an area where consideration of wind condition may provide useful economic
benefits. If the regulators allow it, which may depend on flare quota considerations,
the actual production rate when the compressors were unavailable could be set based
on the measured windspeed and direction for the period in question. In other words
when the windspeed was low or in a beneficial direction the flaring rate could be set at
100% of production. Should this prove attractive to HMDC a set of envelopes for a
range of wind speeds and directions could be prepared which could be used in an
operational procedure to select production rate dependent on wind condition.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
126
9.0 IMPLICATIONS FOR HIBERNIA
9.1 Introduction
In the foregoing sections the various aspects relating the RABS have been analysed.
The intent of this section is to combine the analysis into a form that can be used to
make decisions regarding potential capacity opportunities that exist in the flare system,
as well as identify the issues that will require resolution irrespective of the exercise of
any choices. Generally the potential changes fall into 3 categories:
Capacity opportunities resulting from the application of more modern design
practices (not all of these opportunities add apparent capacity).
Areas where the design documentation should be revised to increase the integrity
and traceability in the design.
Optional changes which can be considered to be related to house keeping.
Therefore this section is separated into three main sections; Firstly an outline of the
capacity opportunities is given including the apparent capacity effects the changes
would have; Secondly, a list and description of the important changes required to
ensure the integrity and traceability of the system design documentation is given;
Lastly, optional changes are described which will aid the future maintenance and
understanding of the design in future years.
Finally, a list of items that do not easily fall into the previous sections is included for
completeness.
In Appendix 2 a proposed scope of work is included which defines, in more
prescriptive terms, the work required to revalidate the flare system design assuming
HMDC decide to implement the changes described in this section in their entirety.
9.2 Flare System Capacity Opportunities
The following summarises the capacity opportunities available in the flare system with
respect to new codes and best practices.
8266-HIB-TN-C-0001 Page of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
127
Table 9.26 Effect of Changing Flare System Design Philosophy on the Apparent Design Capacity (Total Blowdown)
Issue Description /
concern
Case Meets current
code req's?
Safety Cost Failure potential Flare system
capacity*
Recommendation
Jet Fire Described in Section
6.2. Are vessels
sufficiently protected
from the effects of jet
fire?
Original design did not
purposefully include
mitigating measures
Design
Safety analysis not carried
through to engineering.
Codes do not
require
measures to be
included for jet
fire (API
currently
working to
change this)
N/A Rapid escalation. None Adopt best practice.
Ensure insulation
integrity on lower
pressure systems
during jet flame impulse
momentum.
Best practice
Detailed 3D analysis and
prevention measures to
ensure vessel will not fail in
a jet fire
Rapid escalation prevented
unless insulation fails.
Reducing
blowdown
start
pressure
Described in Section
6.3. Compressors
blowdown from PSHH
setting. Rest of
system depressures
from normal pressure.
Design Exceeds code
requirement
+ N/A N/A No effect on HP flare.
LP flare capacity
available is increased
by ~ 17,000 kg/h
compared to the
original blowdown case
89,601 kg/h
HMDC have declined
this change for now,
preferring the more
conservative design
approach (which avoids
changes to blowdown
calculations should
compressor operating
conditions change
significantly). The
capacity opportunity will
be described in an
appendix in the updated
RABS.
Best practice
As blowdown initiates
automatically, design
system with normal start
pressure
In the unlikely event that there
was a fire coincident with a shut
in situation (that was not caused
by trip) the blowdown rate would
be higher than anticipated and if
the wind were adverse could
lead to higher than planned
radiation levels on the platform.
Reducing
the
Described in Section
6.3 Certain vessels
Design Exceeds RP 521
requirements in
N/A N/A No significant effect on
LP flare because the
Best practice would
require all blowdown
8266-HIB-TN-C-0001 Page 128 of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
Issue Description /
concern
Case Meets current
code req's?
Safety Cost Failure potential Flare system
capacity*
Recommendation
blowdown
end
pressure
(i.e. with wall
thicknesses over 1”)
can be depressured to
50% of the design
pressure rather than
690 kPag.
Various components are
depressured to either
690 kPag or 50% of the
design pressure
some instances. HP compressor seal oil
system requires the
compressor and
components to
depressure to
atmospheric pressure.
HP flare capacity
available is increased
by ~ 70% compared to
the original blowdown
case 133,316 kg/h
(reduces to
~40,000 kg/h)
calculations to be re-run
and new orifice plates in
the affected blowdown
section. For the
moment best practice is
declined. An appendix
to the updated RABS
will be created to
identify the capacity
opportunity in case it is
required in the future.
Best practice
Systems with design pressure above 1724 kPag should be depressured to 50% of
the design pressure. Systems with design
pressure below
1724 kPag need not be
depressured. However, if
it is chosen to do so, the
final pressure should be
690 kPag or 50% of the
design pressure,
whichever is less.
Vessels with wall
thicknesses below 1 inch
should be considered
separately.
More closely
follows the intent
of RP 521
+ N/A
Blowdown
stagger
Described in Section
6.4. Stagger not
sufficiently
independent and
equipment is in same
fire zone. Fire may
affect A train injection
compressor, yet
blowdown will be held.
Design Ambiguous
(although the
recommended
practices allow
controlled
blowdown)
1. Stagger fails closed would
lead to escalation.
2. Stagger fails open at
initiation leads to high radiation
levels.
3. Jet fire analysis suggests
injection compressor vessels
should not fail.
33,974 kg/h of load is
added to the LP flare
blowdown case.
(compared to
89,601 kg/h original LP
flare blowdown design
case. New rate is
therefore
Decline best practice.
The A train injection
compressor
components are not at
risk due to their
thickness.
QRA the software
reliability.
8266-HIB-TN-C-0001 Page 129 of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
Issue Description /
concern
Case Meets current
code req's?
Safety Cost Failure potential Flare system
capacity*
Recommendation
123,575 kg/h).Best practice avoids stagger
unless the systems can be
made sufficiently
independent
+ 1. Difficult problem relating to
back pressure on LP separator
to overcome.
2. Otherwise, very reliable.
Higher radiation levels designed
for.
Design
windspeed
and direction
Described in Section
6.6. The design
windspeed is higher
than absolutely
necessary.
Design = 27 m/s from North No code
requirements.
N/A If windspeed is higher and
design release is occurring the
radiation levels on the platform
will be exceeded.
HP and LP flare
apparent capacity is
increased by
approximately 7% (i.e.
by 9,000 kg/h and
6,000 kg/h
respectively).
Best practice declined.
For continuous flaring
case a risk mitigation
procedure could be
developed to increase
the flaring rate when the
compressors were down
dependent on the
measured wind
condition.
Best practice = 24.5 m/s
from North
34.2 m/s from North West
No code
requirements
If windspeed is higher and
design release is occurring, the
radiation levels for emergency
on the platform will be exceeded
somewhat. It is highly unlikely
that anyone would be on deck
without the necessary protection
in such a case.
Acceptable
flare
radiation
levels
Described in Section
6.7. The requirements
in the RABS are over
conservative.
Design
6.3 kW/m2 at crown block and 3.2 kW/m2 at escape ways
Over
conservative
N/A N/A HP and LP flare capacity is increased by approximately 50% before new radiation levels are approached during blowdown (i.e. by 60,000 kg/h and 45,000 kg/h for the HP and LP flares respectively).
Adopt and describe best practice in flare documentation. The practice will remove inconsistency compared to Canadian regulations and international standards. Hydraulic considerations may not allow the full use of new capacity.
Best practice - Radiation levels raised to:
9.5 kW/m2 at crown block (shielded)
6.3 kW/m2 at any escape way (no shielding)
3.2 kW/m2 at the helideck
1.9 kW/m2 continuous at the weather deck
(and meets
Canadian
regulations)
N/A
8266-HIB-TN-C-0001 Page 130 of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
Issue Description /
concern
Case Meets current
code req's?
Safety Cost Failure potential Flare system
capacity*
Recommendation
Incorporate
the effects of
vessel
insulation on
the vapour
rates
Described in Section
6.9.1.
Design
No credit taken for insulation on vessels in blowdown calculations
Over
conservative
N/A N/A HP and LP flare capacity is increased by approximately 5% during blowdown case (i.e. by 6,000 kg/h and 4,000 kg/h for the HP and LP flares respectively).
Best practice would require all blowdown calculations to be re-run. For the moment best practice is declined. A note will be incorporated in the revised flare documents to note the capacity opportunity in case it is required in the future.
Best practice
Credit taken for insulation
N/A
Remove the
effect of
future
equipment
Described in Section
7.0.
N/A N/A N/A N/A N/A HP flare capacity is increased by approximately 38,000 kg/h during blowdown. LP flare system capacity unchanged.
Incorporate the revised data in the updated RABS. Identify the “spare” capacity for future projects in a suitable section.
N/A N/A N/A Negl. N/A
= acceptable - + = most acceptable
= least expensive - = most expensive
* by making change to best practice
8266-HIB-TN-C-0001 Page 131 of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
Care needs to be exercised when considering Table 9.26 as the values are not
additive. What the table does show however is the significant spare capacity in the HP
flare when considering the blowdown cases. The LP flare is very different. The
changes that affect capacity alone (rather than implied through radiation calculations)
are insufficient to offset the large change required to remove the blowdown stagger.
Therefore this change would force the design rate of the system to be increased and
would therefore require detailed hydraulic analysis to be undertaken. The difficult
aspect will be the superimposed back pressure on the LP separator. If this is too high
there will be the undesirable consequence of raising the pressure in the LP separator
when the blowdown valve opens. This would have the effect of increasing inventory
and reducing the time to failure of the LP separator if exposed to fire. Some mitigating
measures would likely be required in this instance. Given this and the apparent
inherent capability of the injection compressor components to survive the pause period
before blowdown commences, suggests that the stagger in the system should be
retained.
For the relief cases other than blowdown, two out of three of the capacity opportunities
(i.e. relief cases which were overestimated during design) are negative. The worst of
the problems relates to the spillover valve failure cases as these have the potential to
significantly exceed the HP flare system design rate. This is shown on Table 9.27.
which follows.
8266-HIB-TN-C-0001 Page 132 of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
Table 9.27 Effect of Changing Flare System Design Philosophy on the Apparent Design Capacity (Relief Cases)
Issue Description /
concern
Case Meets current
code req’s?
Safety Cos
t
Failure potential Flare system capacity* Recommendation
Two-phase
relief (See
Section 5.3
and 6.5)
New sizing
method
increases valve
size required for
this case
Design - Old additive API method No longer N/A If the design case is the
sizing case the vessel can
be overpressured.
The design rate which can
be accommodated in the
existing valves reduces
dramatically. For
comparison max single well
rate for HP separator is ~
54 kbopd.
Adopt best practice.
This issue requires the
maximum well rate and
maximum number of
wells to be redefined as
they are likely to
compromise the RV size
on the HP separator.
The test separator RVs
are being replaced.
Best practice - New API method
Valve size may be to high
and valve will chatter.
Missed relief
cases (See
Section 5.2)
Failure of
spillover valves
(open) exceeds
flare system
capacity.
Design - missed a valid relief case No. N/A If valve fails open the flare
system design rate is
significantly exceeded.
Flare system capacity was
not dimensioned for the
dimensioning case.
Adopt best practice and
use measures to limit
peak load.
Best practice - Design for any single
valve failure.
N/A
Blowby
cases
methodology
flawed (See
Section
5.3.1)
Blowby cases
are over
conservative.
This would
prevent the
installation of
larger LCVs if
this proved
necessary.
Design - Over conservative case
assumed.
N/A Valve is likely to chatter if
faced with the blowby
case.
As there will be no desire to
change the existing valve,
there will be a latent capacity
in the system which can be
used for future upgrades.
The capacity change
available (which would
translate to an increase in
separator LCV size) is
approximately 20%.
Add note to RABS
update to describe the
spare capacity.
Best practice:
Use settleout pressure for
shutdown case
Take account of downstream
control valve positions and fluid
properties in production case.
N/A A valve, properly sized for
the case in question, will
not chatter when faced
with the design case.
= acceptable - + = most acceptable
= least expensive - = most expensive
* by making change to best practice
8266-HIB-TN-C-0001 Page 133 of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
In the above tables the issues which add capacity are optional. In other words these
are issues that HMDC can adopt or decline at will. The issues which have a negative
capacity effect, for obvious reasons, will require some work to resolve.
9.3 Impact on the Design Documentation
The outcome of the technical audit indicates the following aspects of the design will
need to be revisited / revised. The table is a significantly shortened version of the
table presented in Section 5.4 and represents the most important changes that should
be considered. Repetitive items (including those in tables 9.1 and 9.2) and issues
requiring simple comment in the RABS update are omitted. The table should be read
in this light.
8266-HIB-TN-C-0001 Page 134 of 139 Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc October 2000
Audit Tasks Methodology Consistency X See /1 As Built Key Assumptions
HP Flare system isometric Rev A1
Considers pressure at end of sub-header (i.e. main header pressure) is atmospheric (i.e. that this is a singular event not
coincident with any other releases)
Maximum allowable built-up back pressure at PSV discharge must be < 10% of set pressure i.e. conventional valves - lots of margin
- (suitability of conventional valves to be confirmed by inst / vendor - datasheet at rev C2, 11-Nov-92)
Relief Case Rate Configuration
Backflow 23,334 kg/h 1 x 100% operation
PSV datasheet States 10% accumulation but 'Max Relieving Pressure' given is 121% of set pressure
ESI compressible flow analysis sufficient
Key Results
Maximum Mach No. in system = 0.56
Line sizes sufficient
Relief Valve Datasheet backpressure
Calculated Backpressure
33-PSV- kPa(a) kPa(a)
7326 102-601 174
Issues
34-036/1 Inconsistency on datasheet between accumulation and 'Max Relieving Pressure' (should be 10%)
Basis for backflow calculation not given (probably NRV failure)
Possible that HP flare isometric was out of date when calc was made however unlikely to effect results significantly (line sizes
used are correct compared to 'As Built' P&ID)
Audited by AJR Date 21-Jun-00 Checked by MFG Date 15-Aug-00
PSV Equipment Protected
33-PSV-7326 Inj Stage Suction Scrubber B
33-PSV-7226 3rd Stage Suction Scrubber B
PSV Equipment Protected
8266-HIB-TN-C-0001 Appendix I Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc Page 26 of 42 October 2000
Calculation Description Rev DateNumber /
Calculation Book
34-037 / G HM & CM Expansion Drums Simultaneous Fire Relief Case - Network Analysis 02 01-Feb-93
Audit Tasks Methodology Consistency X See /1 As Built Key Assumptions
LP Flare system isometric Rev A3
Considers pressure at end of sub-header (i.e. main header pressure) is atmospheric (i.e. that this is a singular event not
coincident with any other releases)
Maximum allowable built-up back pressure at PSV discharge must be < 10% of set pressure (i.e. conventional valves)
Equipment Relief Case Rate Configuration
Protected
CM Exp'n Drum Fire 1,055 kg/h 1 x 100% operation + 1 x 100% Standby
HM Exp'n Drum Fire 27,628 kg/h 1 x 100% operation + 1 x 100% Standby
ESI compressible flow analysis sufficient
Key Results
Maximum Mach No. in system = 0.45
Line size increase for common line from 6" to 8". Other line sizes sufficient
Datasheet backpressure
Calculated Backpressure*
kPa(a) kPa(a) * incorporating line size increase
102-136 136
102-136 185
Issues
34-037/1 Calculated backpressure (for 0152A/B) greater than specified on datasheet - calc considers
this OK as less than 10% of set pressure
Possible that LP flare isometric was out of date when calc was made however unlikely to effect results significantly (line sizes
used are correct compared to 'As Built' P&ID)
Audited by AJR Date 21-Jun-00 Checked by MFG Date 15-Aug-00
PSV
64-PSV-0118A/B
63-PSV-0152A/B
64-PSV-0118A/B
63-PSV-0152A/B
Relief Valve
8266-HIB-TN-C-0001 Appendix I Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc Page 27 of 42 October 2000
Calculation Description Rev DateNumber /
Calculation Book
34-044 / G Total LP Blowdown - After 3 mins (stagger point) - Network Analysis 01 22-Mar-93
Audit Tasks Methodology Consistency As Built Key Assumptions
LP Flare system isometric Rev A3
Blowdown simulations HIBBD200/202/203/154.LIS
Total blowdown rate after 3 mins (stagger point) = 85,504 kg/h
Staggered flow - 3 minute time delay on Inj 'A' Compressor / Scrubber
Compositions from blowdown simulations referenced above
No tip P (pipe flare)
Key Results
Calculation undertaken to check velocities were acceptable. Maximum Mach No. in system = 0.36 therefore line sizes sufficient
Blowdown Valve Rev D1 Datasheet
backpressure
Network Analysis
Backpressure
33-ESV- kPa(a) kPa(a)
7279 201 176
7350 201 253
7372 201 157
7153 201 193
7177 201 157
7253 201 209
Additional valves on summary sheet not identified and no supporting blowdown valve datasheet attached
Issues
None
Total blowdown rate (initial rate) referenced in calc less than that in Relief & Blowdown Study Report ( 89,601 kg/h) but not
sufficient to effect sizing
Some calculated backpressures greater than specified on blowdown valve Rev D1 datasheet but not sufficient to affect sizing
See also 34-042/2
Possible that LP flare isometric was out of date when calc was made however unlikely to effect results significantly (line sizes
used are correct compared to 'As Built' P&ID)
Audited by AJR Date 21-Jun-00 Checked by MFG Date 15-Aug-00
Calculation Description Rev DateNumber /
Calculation Book
34-046 / G Fuel Gas Cooler / Heater tube rupture relief line size check 01 02-Mar-93
Audit Tasks Methodology N/A Consistency N/A As Built X See /1
Issues
34-046/1 'As Built' P&IDs show bursting discs in this service (calc considers PSVs) therefore calc is no longer valid
Audited by AJR Date 21-Jun-00 Checked by MFG Date 15-Aug-00
8266-HIB-TN-C-0001 Appendix I Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc Page 28 of 42 October 2000
Calculation Description Rev DateNumber /
Calculation Book
34-047 / G Simultaneous Fire Relief Case from E-6202 & Z-6201 A/B 01 02-Mar-93
Audit Tasks Methodology Consistency As Built Key Assumptions
HP Flare system isometric No Rev Given
Considers pressure at end of sub-header (i.e. main header pressure) is atmospheric (i.e. that this is a singular event not
coincident with any other releases)
62-PSV-0040A/B stated to be conventional valves yet 'Normal Back Pressure' is 1-500 kPa(g) (i.e. > 10% set pressure)
- others are balanced for similar 'Normal Back Pressure'
PSV Relief Case Rate Configuration
62-PSV-
0040A/B Fire 1,346 kg/h 1 x 100% operation + 1 x 100% Standby
0081 Fire 1,814 kg/h 1 x 100% operation
0092 Fire 1,814 kg/h 1 x 100% operation
ESI compressible flow analysis sufficient
Key Results
Maximum Mach No. in system = 0.56
Header 3"-VH-34205 increased in size (shown as 3"-VH-34326 & 4"-VH-34327 on 'As Built' P&ID). Original size gave
mach no. = 0.97
Other line sizes sufficient
Relief Valve Datasheet backpressure
Calculated Backpressure*
62-PSV- kPa(a) kPa(a) * incorporating line size increase
0040A/B 102-601 153
0081 102-601 156
0092 102-601 160
Issues
None
Possible that HP flare isometric was out of date when calc was made however unlikely to effect results significantly (line sizes
used are correct compared to 'As Built' P&ID)
Audited by AJR Date 22-Jun-00 Checked by MFG Date 15-Aug-00
Calculation Description Rev DateNumber /
Calculation Book
34-048 / G 2nd Stage Suction Scrubber A (D-3302A) PSV Discharge Line Size Confirmation 01 02-Mar-93
Audit Tasks Methodology Consistency As Built Key Assumptions
HP Flare system isometric No Rev Given
Considers pressure at end of sub-header (i.e. main header pressure) is atmospheric (i.e. that this is a singular event not
coincident with any other releases)
33-PSV-7099 balanced valve - suitability of balanced valve to be confirmed by inst / vendor - datasheet at rev C2, 11-Nov-92
Relief Case Rate Configuration
Backflow 5,080 kg/h 1 x 100% operation
ESI compressible flow analysis sufficient
Key Results
Maximum Mach No. in system = 0.53
Line sizes sufficient
Relief Valve Datasheet backpressure
Calculated Backpressure
33-PSV- kPa(a) kPa(a)
7099 102-601 157
Issues
None
Basis for backflow calculation not given (probably NRV failure)
Possible that HP flare isometric was out of date when calc was made however unlikely to effect results significantly (line sizes
used are correct compared to 'As Built' P&ID)
Audited by AJR Date 22-Jun-00 Checked by MFG Date 15-Aug-00
Z-6201B
PSV Equipment Protected
33-PSV-7099 D-3302A
Equipment Protected
E-6202 (Tube side)
Z-6201A
8266-HIB-TN-C-0001 Appendix I Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc Page 29 of 42 October 2000
Calculation Description Rev DateNumber /
Calculation Book
34-049 / G Inj Stage Suction Scrubber A (D-3304A) PSV Discharge Line Size Confirmation 01 02-Mar-93
Audit Tasks Methodology Consistency As Built Key Assumptions
HP Flare system isometric No Rev Given
Considers pressure at end of sub-header (i.e. main header pressure) is atmospheric (i.e. that this is a singular event not
coincident with any other releases)
33-PSV-7301 conventional valve - suitability of conventional valve to be confirmed by inst / vendor - datasheet at rev C2, 11-Nov-92
Relief Case Rate Configuration
Backflow 23,334 kg/h 1 x 100% operation
ESI compressible flow analysis sufficient
Key Results
Maximum Mach No. in system = 0.59
Line sizes sufficient
Relief Valve Datasheet backpressure
Calculated Backpressure
33-PSV- kPa(a) kPa(a)
7301 102-601 188
Issues
None
Basis for backflow calculation not given (probably NRV failure)
Possible that HP flare isometric was out of date when calc was made however unlikely to effect results significantly (line sizes
used are correct compared to 'As Built' P&ID)
Audited by AJR Date 22-Jun-00 Checked by MFG Date 15-Aug-00
Calculation Description Rev DateNumber /
Calculation Book
34-050 / G 3rd Stage Suction Scrubber A (D-3303A) PSV Discharge Line Size Confirmation 01 02-Mar-93
Audit Tasks Methodology Consistency As Built X See /1
Key Assumptions
HP Flare system isometric No Rev Given
Considers pressure at end of sub-header (i.e. main header pressure) is atmospheric (i.e. that this is a singular event not
coincident with any other releases)
33-PSV-7200 conventional valve - suitability of conventional valve to be confirmed by inst / vendor - datasheet at rev C2, 11-Nov-92
Relief Case Rate Configuration
Backflow 13,233 kg/h 1 x 100% operation
ESI compressible flow analysis sufficient
Key Results
Maximum Mach No. in system = 0.60
Line sizes sufficient
Relief Valve Datasheet backpressure
Calculated Backpressure
33-PSV- kPa(a) kPa(a)
7200 102-601 229
Issues
34-050/1 Rev C2 PSV datasheet states set pressure = 8200 kPa(g), 'As Built' P&ID shows set pressure = 7000 kPa(g)
Basis for backflow calculation not given (probably NRV failure)
Possible that HP flare isometric was out of date when calc was made however unlikely to effect results significantly (line sizes
used are correct compared to 'As Built' P&ID)
Audited by AJR Date 22-Jun-00 Checked by MFG Date 15-Aug-00
33-PSV-7301 D-3304A
PSV Equipment Protected
PSV Equipment Protected
33-PSV-7200 D-3303A
8266-HIB-TN-C-0001 Appendix I Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc Page 30 of 42 October 2000
Calculation Description Rev DateNumber /
Calculation Book
34-051 / G HP Fuel gas KO Drum (D-6201) PSV Discharge Line Size Confirmation 01 02-Mar-93
Audit Tasks Methodology Consistency As Built Key Assumptions
HP Flare system isometric No Rev Given
Considers pressure at end of PSV discharge line (i.e. sub-header pressure) is atmospheric (i.e. that this is a singular event not
coincident with any other releases)
62-PSV-0024A/B balanced valves
PSV Relief Case Rate Configuration
62-PSV-
0024A/B Fire 10,895 kg/h 1 x 100% operation + 1 x 100% Standby
ESI compressible flow analysis sufficient
Key Results
Maximum Mach No. in system = 0.49
Line sizes sufficient
Relief Valve Datasheet backpressure
Calculated Backpressure
62-PSV- kPa(a) kPa(a)
0024A/B 102-601 210
Issues
See 34-033/1
Possible that HP flare isometric was out of date when calc was made however unlikely to effect results significantly (line sizes
used are correct compared to 'As Built' P&ID)
Audited by AJR Date 22-Jun-00 Checked by MFG Date 15-Aug-00
Calculation Description Rev DateNumber /
Calculation Book
34-052 / G E-3301 Shell Side PSV Discharge Line Size Confirmation 01 02-Mar-93
Audit Tasks Methodology Consistency As Built Key Assumptions
LP Flare system isometric No Rev Given
Considers pressure at end of PSV discharge line (i.e. sub-header pressure) is atmospheric (i.e. that this is a singular event not
coincident with any other releases)
33-PSV-0094 conventional valve - suitability of conventional valve to be confirmed by inst / vendor - datasheet at rev C2, 20-Nov-92
Relief Case Rate Configuration
Fire 1,182 kg/h 1 x 100% operation
ESI compressible flow analysis sufficient
Key Results
Maximum Mach No. in system = 0.27
Line sizes sufficient
Relief Valve Datasheet backpressure
Calculated Backpressure
33-PSV- kPa(a) kPa(a)
0094 102-136 115
Issues
None
Possible that LP flare isometric was out of date when calc was made however unlikely to effect results significantly (line sizes
used are correct compared to 'As Built' P&ID)
Audited by AJR Date 22-Jun-00 Checked by MFG Date 15-Aug-00
33-PSV-0094 E-3301
PSV Equipment Protected
Equipment Protected
D-6201
8266-HIB-TN-C-0001 Appendix I Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc Page 31 of 42 October 2000
Calculation Description Rev DateNumber /
Calculation Book
34-053 / G E-3303B Shell Side PSV Discharge Line Size Confirmation 01 02-Mar-93
Audit Tasks Methodology N/A Consistency N/A As Built X See below
Issues
None
'As Built' P&IDs show bursting discs installed in this service (calc considers PSVs) therefore calc is no longer valid
See also 34-046/1
Audited by AJR Date 22-Jun-00 Checked by MFG Date 15-Aug-00
Calculation Description Rev DateNumber /
Calculation Book
34-054 / G HP Manifold Relief - Network Analysis 01 02-Mar-93
Audit Tasks Methodology Consistency As Built X See /1
Key Assumptions
HP Flare system isometric No Rev Given
Conventional valve
Calculation takes into account total system pressure drop but still assumes that this is a singular event not coincident
with any other releases
PSV Relief Case Rate Configuration
31-PSV-
7042A/B Fire 74,585 kg/h 1 x 100% operation + 1 x 100% Standby
Tip P estimated from Kaldair supplied graph
ESI compressible flow analysis sufficient
Key Results
Maximum Mach No. in system = 0.68
Line sizes sufficient
Calculated tip P = 33 kPa
Relief Valve Datasheet backpressure
Calculated Backpressure
31-PSV- kPa(a) kPa(a)
7042A/B 102-601 226
Issues
34-054/1 Rev C2 PSV datasheet states set pressure = 34,400 kPa(g), 'As Built' P&ID shows set pressure = 34,100 kPa(g)
Possible that HP flare isometric was out of date when calc was made however unlikely to effect results significantly (line sizes
used are correct compared to 'As Built' P&ID)
Audited by AJR Date 22-Jun-00 Checked by MFG Date 15-Aug-00
Equipment Protected
HP Manifold
8266-HIB-TN-C-0001 Appendix I Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc Page 32 of 42 October 2000
Calculation Description Rev DateNumber /
Calculation Book
34-055 / G Simultaneous Fire Relief Case from Z-3701 A/B, Z-3702 A/B & Z-6202 A/B 01 10-Feb-93
Audit Tasks Methodology Consistency As Built Key Assumptions
HP Flare system isometric No Rev Given
Balanced valves
Calculation takes into account total system pressure drop but still assumes that this is a singular event not coincident
with any other releases
Relief Case Rate Configuration
Fire 17,345 kg/h 1 x 100% operation
Fire 12,688 kg/h 1 x 100% operation
Fire 17,345 kg/h 1 x 100% operation
Fire 12,688 kg/h 1 x 100% operation
Fire 1,814 kg/h 1 x 100% operation
Fire 1,814 kg/h 1 x 100% operation
Tip P estimated from Kaldair supplied graph
ESI compressible flow analysis sufficient
Key Results
Maximum Mach No. in system = 0.39
Line sizes sufficient
Calculated tip P = 50 kPa
Relief Valve Datasheet backpressure
Calculated Backpressure
37-PSV- kPa(a) kPa(a)
1021 102-601 344
1043 102-601 307
1063 102-601 336
1001 102-601 313
62-PSV-
0111 102-601 285
0122 102-601 282
Issues
See 34-033/1
Possible that HP flare isometric was out of date when calc was made however unlikely to effect results significantly (line sizes
used are correct compared to 'As Built' P&ID)
Audited by AJR Date 23-Jun-00 Checked by MFG Date 15-Aug-00
37-PSV-1043
Z-3702A
Z-3701B
62-PSV-0122
Z-6202A
37-PSV-1063
Z-6202B
Z-3702B
62-PSV-0111
Z-3701A37-PSV-1001
PSV Equipment Protected
37-PSV-1021
8266-HIB-TN-C-0001 Appendix I Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc Page 33 of 42 October 2000
Calculation Description Rev DateNumber /
Calculation Book
34-056 / G Individual HP Separator Blowdown Case - Line Size Confirmation 01 10-Feb-93
Audit Tasks Methodology Consistency As Built Key Assumptions
HP Flare system isometric No Rev Given
Total blowdown rate (initial rate) = 41,311 kg/h
Tip P estimated from Kaldair supplied graph
ESI compressible flow analysis sufficient
Key Results
Maximum Mach No. in system = 0.66
Line sizes sufficient
Calculated tip P = 33 kPa
Blowdown Valve Rev D1 Datasheet
backpressure
Calculated Backpressure
31-ESV- kPa(a) kPa(a)
7318 551 402
Issues
None
Possible that HP flare isometric was out of date when calc was made however unlikely to effect results significantly (line sizes
used are correct compared to 'As Built' P&ID)
Audited by AJR Date 23-Jun-00 Checked by MFG Date 15-Aug-00
Calculation Description Rev DateNumber /
Calculation Book
34-057 / G E-3701 Shell & Tube Side Simultaneous Fire Relief Case - Line Size Confirmation 01 12-Feb-93
Audit Tasks Methodology Consistency X See /1 As Built Key Assumptions
LP Flare system isometric No Rev Given
Considers pressure at end of sub-header (i.e. main header pressure) is atmospheric (i.e. that this is a singular event not
coincident with any other releases)
Conventional valvesEquipment Protected Set Pressure Relief Case Rate Configuration
E-3701 (SS) 1380 kPag Fire 15,804 kg/h 1 x 100% operation
E-3701 (TS) 780 kPag Fire 2,135 kg/h 1 x 100% operation
ESI compressible flow analysis sufficient
Key Results
Maximum Mach No. in system = 0.48
Calculation considers that line sizes sufficient
Relief Valve Datasheet backpressure
Calculated Backpressure
37-PSV- kPa(a) kPa(a)
1482 102-136 197
1497 102-136 161
Issues
34-057/1 Calculated backpressure exceeds that specified on datasheet for both PSVs
See also 34-033/1
Possible that LP flare isometric was out of date when calc was made however unlikely to effect results significantly (line sizes
used are correct compared to 'As Built' P&ID)
Audited by AJR Date 23-Jun-00 Checked by MFG Date 15-Aug-00
37-PSV-1497
PSV
37-PSV-1482
8266-HIB-TN-C-0001 Appendix I Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc Page 34 of 42 October 2000
Calculation Description Rev DateNumber /
Calculation Book
34-058 / G E-6201A/B Tube Side Fire Relief Case - Line Size Confirmation 01 10-Feb-93
Audit Tasks Methodology Consistency As Built Key Assumptions
HP Flare system isometric No Rev Given
Considers pressure at end of PSV discharge line (i.e. sub-header pressure) is atmospheric (i.e. that this is a singular event not
coincident with any other releases)
Conventional valves
PSV Relief Case Rate Configuration
62-PSV-
0001A/B Fire 828 kg/h 1 x 100% operation + 1 x 100% Standby
ESI compressible flow analysis sufficient
Key Results
Maximum Mach No. in system = 0.22
Line sizes sufficient
Relief Valve Datasheet backpressure
Calculated Backpressure
62-PSV- kPa(a) kPa(a)
0001A/B 102-601 112
Issues
See 34-033/1
Possible that HP flare isometric was out of date when calc was made however unlikely to effect results significantly (line sizes
used are correct compared to 'As Built' P&ID)
Audited by AJR Date 23-Jun-00 Checked by MFG Date 15-Aug-00
Calculation Description Rev DateNumber /
Calculation Book
34-059 / G Comparative Program check of INPLANT Single Phase Simulation vs ESI 01 23-Apr-93
Audit Tasks Methodology X See /1 Consistency As Built Key Assumptions
HP Separator Max Spill-off Case
Key Results
Pressure Drop given by ESI run ~20% less than INPLANT
Velocity given by ESI run 5% max less than INPLANT
Mach Nos. given by ESI run ~10% max less than INPLANT
Probably due to estimated average fluid properties used in ESI runs
Issues
34-059/1 Accuracy of calculations using ESI instead of INPLANT
Audited by AJR Date 23-Jun-00 Checked by MFG Date 15-Aug-00
E-6201A/B (Tube side)
Equipment Protected
8266-HIB-TN-C-0001 Appendix I Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc Page 35 of 42 October 2000
Calculation Description Rev DateNumber /
Calculation Book
34-034 / G 2nd Stage Suction Scrubber B PSV - Network Analysis 02 26-Mar-93
Audit Tasks Methodology Consistency As Built Key Assumptions
HP Flare system isometric Rev A3
Considers pressure at end of sub-header (i.e. main header pressure) is atmospheric (i.e. that this is a singular event not
coincident with any other releases)
33-PSV-7124 balanced valve - suitability of balanced valve to be confirmed by inst / vendor - datasheet at rev C2, 11-Nov-92
Relief Case Rate Configuration
Backflow 5,080 kg/h 1 x 100% operation
ESI compressible flow analysis sufficient
Key Results
Maximum Mach No. in system = 0.51
Line sizes sufficient
Relief Valve Datasheet backpressure
Calculated Backpressure
33-PSV- kPa(a) kPa(a)
7124 102-601 162
Issues
None
Basis for backflow calculation not given (probably NRV failure)
Possible that HP flare isometric was out of date when calc was made however unlikely to effect results significantly (line sizes
used are correct compared to 'As Built' P&ID)
Audited by AJR Date 23-Jun-00 Checked by MFG Date 15-Aug-00
Calculation Description Rev DateNumber /
Calculation Book
34-064 / G HP Separator Max 2-Phase Relief - Network Analysis 01 07-Jun-93
Audit Tasks Methodology Consistency As Built Key Assumptions
HP Flare system isometric Rev A4
HP Separator Blocked Outlet via PSV-7308A. Total load = 252,372 kg/h (based on 1 max well (30,000 bpd)
+ 1 average well (10,000 bpd) flowing)
Note at front of calc states that Rev 7 of Design Basis gives max well flow of 20,000 bpd + average well i.e. 30,000 bpd total
- calc considers 40,000 bpd flowrate anyway
INPLANT separator module not working therefore calc considers 2-phase flow to flare tip
No PSV datasheets included in calculation. Datasheet for PSV-7308 included in Calc 34-022 is for vapour relief only.
Tip P estimated from Kaldair supplied graph based on 64,273 kg/h vapour
Key Results
Maximum Mach No. in system = 0.31
Line sizes appear sufficient
Relief Valve Datasheet backpressure
Calculated Backpressure
31-PSV- kPa(a) kPa(a)
7308 102-851 646
Issues
None
See also PSV calculation technical audit
Audited by AJR Date 23-Jun-00 Checked by MFG Date 15-Aug-00
33-PSV-7124 D-3302B
PSV Equipment Protected
8266-HIB-TN-C-0001 Appendix I Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc Page 36 of 42 October 2000
Book H General
Book H contains a number of 'Check Print' copies of calculations reviewed in earlier volumes. There are nosignificant comments to record. In addition there a number of un-numbered calculation (all superseded). These 'Check Prints' and un-numbered calculations have not been reviewed in detail.
Calculation Description Rev DateNumber /
Calculation Book
34-001 / H Combined LP/HP KO Drum Sizing (Preliminary) No Rev 31-Jan-91
Audit Tasks Methodology N/A Consistency N/A As Built N/A
Calculation superseded. Not reviewed in detail
Audited by AJR Date 26-Jun-00 Checked by MFG Date 15-Aug-00
Calculation Description Rev DateNumber /
Calculation Book
34-002 / H HP KO Drum Sizing (Check) No Rev 04-Feb-91
Audit Tasks Methodology N/A Consistency N/A As Built N/A
Calculation superseded. Not reviewed in detail
Audited by AJR Date 26-Jun-00 Checked by MFG Date 15-Aug-00
Calculation Description Rev DateNumber /
Calculation Book
34-003 / H Flare Boom Length Calculations (Preliminary) No Rev 01-Mar-91
Audit Tasks Methodology N/A Consistency N/A As Built N/A
Calculation superseded. Not reviewed in detail
Audited by AJR Date 26-Jun-00 Checked by MFG Date 15-Aug-00
Calculation Description Rev DateNumber /
Calculation Book
34-004 / H Flare System - Material Balance for Flare UFD 0 25-Mar-91
Audit Tasks Methodology N/A Consistency N/A As Built N/A
Calculation superseded. Not reviewed in detail
Audited by AJR Date 26-Jun-00 Checked by MFG Date 15-Aug-00
8266-HIB-TN-C-0001 Appendix I Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc Page 37 of 42 October 2000
Calculation Description Rev DateNumber
31.35 Relief Valve Calculations - HP Separator C1 Nov '91
Audit Tasks Methodology X see /1,4&5 Consistency X see /2&3 As Built X See below
Audit Tasks Methodology X see /1&2 Consistency As Built Four Cases Considered; Fire, Gas Blowby From MP Separator, Gas Blowby From Test Separators (Individually),
Gas Blowby From 2nd Stage Suction Scrubber
Key Assumptions
Fire Case
Vessel dimensions: 4.2m x 23.2m T/T
No credit taken for vessel insulation
Gas Blowby From MP Separator
Control valve upstream pressure 12.35 bara
Valve CV = 700 max
Two LCVs in parallel installed but only one valve fails at any one time
Gas Blowby From Test Separator
Control valve upstream pressure 12.04 bara
Valve CV = 195 max
Gas Blowby From 2nd Stage Suction Scrubber
Control valve upstream pressure 11.00 bara
Valve CV = 16 max
Key Results
Results Summarised in table below
Case Relief Orifice Area
Flowrate Required
kg/h in2
Fire Case 14,195 3.28
Gas Blowby - MP Separator 110,874 28.80 Governing Case
Gas Blowby - Test Separator 24,853 Not Calc'd
Gas Blowby - 2nd Stage Scrubber 2,384 Not Calc'd
Installed PSV orifice area from 'As Built' datasheet = 2 x 16 in2 operating.
Issues
31.37/1 Is it possible for the Test Separator manifold to be connected to the LP Separator when operating in high pressure
mode?
31.37/2 Are 2 x 50% LCVs sufficiently independent?
See also 31.36/6
Actual installed MP Separator level control valve CV = 600 for blowby case.
Actual installed 2nd Stage Scrubber level control valve CV = 24.0. No concern as this is not a major relief case.
Audited by AJR Date 07-Aug-00 Checked by MFG Date 16-Aug-00
8266-HIB-TN-C-0001 Appendix I Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc Page 40 of 42 October 2000
Calculation Description Rev DateNumber
31.38 Inlet Line Size Checking for Relief Valves 05-Dec-91
Audit Tasks Methodology X see below Consistency X see /1 As Built X see /1
Key Assumptions
Inlet line equivalent length 100m
Preliminary data used for relief loads and selected orifice areas
Key Results
Relief valve inlet line sizes
Issues
31.38/1 Inlet line sizes should have been recalculated using 'Final' relief data and isometrics.
This calculation has obviously been revised as many PSV inlet line sizes are different to those calculated here.
The data used for this calculation is nearly all out of date. Some relief valve tag numbers have changed and all PSV relieving
capacities and maximum relieving capacities are different on the 'As Built' PSV datasheets. Some PSV set pressures are
also different.
Audited by AJR Date 07-Aug-00 Checked by MFG Date 16-Aug-00
8266-HIB-TN-C-0001 Appendix I Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc Page 41 of 42 October 2000
Audit Tasks Methodology Consistency See below As Built See belowEquipment Considered: E-3103 A/B LP Separator Heaters (HM Side)Key AssumptionsOnly fire case consideredExchanger dimensions: 1.75m OD x 7.431m OL (channel length = 1.1m)Wetted area calc takes into account 15m of 12" pipingPSV set pressure 1380 kPagShell is liquid fullLatent heat of vaporisation = 50 Btu/lbKey ResultsRelief flowrate = 40771 kg/h Required orifice area = 2.36 in2IssuesNone. 'As Built' exchanger has slightly different dimensions (1.175m OD x 7.641m OL) to those used in the calculation but considered insignificant.smaller diameter. Installed orifice area = 2.85 in2.Equipment Considered: D-3104 A/B CoalescersKey AssumptionsOnly fire case consideredCoalescer dimensions: 3.05m ID x 9.75 T/TWetted area calc takes based on LSHHPSV set pressure 700 kPagLatent heat of vaporisation = 415 Btu/lb (as per LP Separator)Key ResultsRelief flowrate = 7214 kg/h Required orifice area = 1.61 in2IssuesNone. Installed orifice area = 2.85 in2.Equipment Considered: E-3104 A/B Crude Product Coolers (HC & CM Side)Key AssumptionsOnly fire case consideredExchanger dimensions: 2.61m x 4.4m x 1.18mHC side PSV set pressure 700 kPagCM side PSV set pressure 1500 kPagWetted area calc for HC side uses 50% total surface areaWetted area calc for CM side uses 50% total surface areaLatent heat of vaporisation (HC) = 415 Btu/lb (as per LP Separator)Latent heat of vaporisation (CM) = 817.3 Btu/lb (treated as water)Key ResultsHC Side: Relief flowrate = 1827 kg/h Required orifice area = 0.41 in2CM Side: Relief flowrate = 928 kg/h Required orifice area = 0.15 in2IssuesNone. 'As Built' exchanger has slightly different dimensions (2.877m x 4.4m x 1.359m) to those used in the calculation but considered insignificant. Installed orifice area, HC side = 0.503 in2, CM side = 0.196 in2.Equipment Considered: E-3701 A/B Crude Recirculation Heater (HC & HM Side)Key AssumptionsOnly fire case consideredExchanger dimensions: 0.813m OD x 4.82m OL (channel length = 1.299m)Wetted area calc takes into account 20m for each sideShell is liquid fullHC side PSV set pressure 740 kPagHM side PSV set pressure 1460 kPagWetted area calc for HC (tube) side uses channel surface areaWetted area calc for HM (shell) side uses shell surface areaLatent heat of vaporisation (HC) = 182.1 Btu/lbLatent heat of vaporisation (HM) =50 Btu/lbKey ResultsHC Side: Relief flowrate = 4866 kg/h Required orifice area = 0.33 in2HM Side: Relief flowrate = 15804 kg/h Required orifice area = 1.13 in2IssuesNone. However, HC Side 'As Built' PSV datasheet has relief load = 2135 kg/hr and installed orifice area of 0.785 in2. This isequivalent to using the same properties in this calc as used for the Coalescer and LP Separator calculation (I.e. latent heat of vap. = 415 Btu/lb anf MW= 37.68).There is an error in the calculation as the orifice area req'd calc uses a flowrate in kg/h instead of lb/h. Resulting true requiredorifice area req'd should be = 0.74in2. As the installed orifice area is greater than the true required as given above there is no concern.Equipment Considered: Z-3701 A/B / Z-3702 A/B Crude Oil Pig Launcher / ReceiverKey AssumptionsOnly fire case consideredDimensions Launcher: 547.7 / 706mm ID x 7200mm OL
Receiver: 547.7 / 706mm ID x 11400mm OL Wetted area calc takes into account 15m of 12" pipingPSV set pressure 4500 kPagEquipment is liquid fullLatent heat of vaporisation = 50 Btu/lbKey ResultsLauncher: Relief flowrate = 12688 kg/h Required orifice area = 0.48 in2Receiver: Relief flowrate = 17345 kg/h Required orifice area = 0.66 in2IssuesNone. 'As Built' equipment has slightly different dimensions to those used in the calculation but considered insignificant.Installed orifice area = 0.785 in2 for both pieces of equiment.
Audited by AJR Date 07-Aug-00 Checked by MFG Date 16-Aug-00
8266-HIB-TN-C-0001 Appendix I Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc Page 42 of 42 October 2000
Calculation Description Rev DateNumber
31.43 Gas Blowby (Checking Capacity of Downstream System for Gas Blowby 22-Nov-92
from HP to MP Separator and MP to LP Separator)
Audit Tasks Methodology X see /1-3 Consistency See below As Built See below
Gas Blowby From HP to MP Separator
Key Assumptions
MP Separator PSVs orifice area is 2 x 16 in2 operating.
MP Separator spill off valve CV = 600
MP Separator spill off valve capacity during blowby case is 94,500 kg/h
Both 50% upstream LCVs fail at the same time
Key Results
Total relief capacity of PSVs and spill-off valve operating together is 365,594 kg/h
If HP Separator level control valves have a CV of 329.8 each, the total relief load if both valves fail open (365,594 kg/h) can
be handled by the installed PSVs and spill off valve operating together.
Issues
31.43/1 This calculation considers both upstream LCVs fail open simultaneously. This scenario is not considered in the
Relief & Blowdown Study Report Rev C1 (or in any other calculations reviewed), nor is the platform designed
for its affects.
Note that the 'As Built' HP Separator LCVs CV = 330 each , 'As Built' spill-off valve CV = 600 and 'As Built' MP Separator PSVs
orifice area is 2 x 16 in2 operating.
The maximum relief load given above (365,594 kg/h) is not considered in the Relief & Blowdown Study Report Rev C1 as
the governing HP flare relief case (HP Separator blocked outlet @ 244,897 kg/h) and is not considered in the flare hydraulic calculations
Gas Blowby From MP to LP Separator
Key Assumptions
LP Separator PSVs orifice area is 2 x 16 in2 operating.
LP Separator spill off valve CV = 4145
LP Separator spill off valve capacity during blowby case is 86,713 kg/h
Both 50% upstream LCVs fail at the same time
Key Results
Total relief capacity of PSVs and spill-off valve operating together is 194,875 kg/h
If MP Separator level control valves have a CV of 550 each, the total relief load if both valves fail open (194,875 kg/h) can
be handled by the installed PSVs and spill off valve operating together.
Issues
31.43/2 This calculation considers both upstream LCVs fail open simultaneously. This scenario is not considered in the
Relief & Blowdown Study Report Rev C1 (or in any other calculations reviewed), nor is the platform designed
for its affects.
31.43/3 The calculation identifies the failure of the spillover valve (open) could lead to a relief rate which is higher than
the current design.
Note that the 'As Built' LP Separator LCVs CV = 600 each , 'As Built' spill-off valve CV = 4145 and 'As Built' MP Separator PSVs
orifice area is 2 x 16 in2 operating. 'As Built' total relief load if calculated using the same method as given here will be higher
than 194,875 kg/h as the installed LCV CV = 600 (not 550).
The maximum relief load given above (194,875 kg/h) is not considered in the Relief & Blowdown Study Report Rev C1 as
the governing LP flare relief case (MP to LP Separator gas blowby @ 110,874 kg/h) and is not considered in the
flare hydraulic calculations
Audited by AJR Date 07-Aug-00 Checked by MFG Date 16-Aug-00
APPENDIX II
STAGE 2 PROPOSAL
Flare System Revalidation Study - Stage 2 Proposal Rev C (27 pages)
8266-HIB-TN-C-0001 Appendix II Revision: B/tt/file_convert/55276c7749795994178b46d5/document.doc Page 1 of 1 October 2000