ACI Committee 376 - Concrete

ACI 376 Committee Concrete Structures for RLG Containment

ACI 376 – TG Houston 2/14/2008 Meeting Minutes 1

ACI Committee 376 Concrete Structures for

Refrigerated Liquefied Gas (RLG) Containment

TG Houston Meeting Minutes

Task Group Meeting

Tuesday, March 18, 2008

9 AM – 5:00 PM ExxonMobil Development Company

Room: GP-6 Room 965

ATTENDANCE Voting Members: Neven Krstulovic, Chairman Allen, J. Brannan, M Hjorteset, K.

Associate Members: Ballard, T.

1. CALL TO ORDER

The meeting was called to order by Chairman Krstulovic. 5. ANNOUNCEMENTS No announcements were made at this meeting 2. WORK ON COMMITTEE DOCUMENTS For voting: Junius Allen arrived after discussions were completed and voting was recorded. Chapter 9 – Minimum Performance Criteria • R9.4.1 – Hoff’s comment on inspection of grout ducts was addressed and considered persuasive. The TG believes that

clarification of the commentary to state that, “inspection for blockage should be carried out before, during and after concrete placement, and before grouting”, would result in less confusing wording. The TG voted unanimously to adopt this editorial change (Affirmative: Krustulovic, Brannan, Hjorteset, and Ballard; Absent: Allen).



Chapter 6 – Analysis and Design • Section 6.1.1 – Hoptay’s comment on availability of the Eurocode 2 to engineers was addressed and was considered

persuasive. The original reference that provides the basis for the Eurocode 2 concrete model should be provided in commentary. However, the reference to Eurocode 2 in the normative section should be retained. Also, the full reference to Eurocode 2 will be provided in reference section. This change is considered editorial only and the TG voted unanimously to adopt the change (Affirmative: Krustulovic, Brannan, Hjorteset, and Ballard; Absent: Allen). References: CEB-FIP Model Code 1990, “Material Properties,” Stress-strain and stress Crack Opening Relations. 1993. ENV 1992-1-1, Eurocode 2: Design of concrete structures – Part 1: General rules and rules for buildings.

In addition, many widely available FE computer programs, such as, ADINA and ABAQUS contain this model.

• Section R6.1.2.1 – The TG believes that Mash’s comment regarding removing the word “static non-linear analysis” from the second paragraph is not correct in that pushover analyses are static, however, a clarification of this wording was proposed. Mash’s editorial comment: to include the terms “demonstrated explicitly” were accepted. The second paragraph will be changed from “A force reduction factor R>1 may be used if it can be shown, by means of dynamic or static nonlinear analyses…” to “A force reduction factor R>1 may be used if it can be shown explicitly, by means of dynamic or static nonlinear analyses (e.g., pushover)…”. This change is considered editorial only and the TG voted unanimously to adopt the change (Affirmative: Krustulovic, Brannan, Hjorteset, and Ballard; Absent: Allen). Wu’s comments is persuasive, however, the TG recognizes that certain situations will arise where non-linear analysis will be required for OBE, such as for seismic isolation. However, the TG believes that the first paragraph should be clarified by stating: “In general, linear analysis is used in the case of low seismic regions or OBE, while non-linear analysis is used in regions with higher seismicity and/or SSE. This change is considered editorial only and the TG voted unanimously to adopt the change (Affirmative: Krustulovic, Brannan, Hjorteset, and Ballard; Absent: Allen).

Chapter 10 – Commissioning • Section 10.2.X – Negative votes by Hoff and Hoptay were addressed and considered persuasive. The TG suggested the

following wording for the first paragraph: “Where anchorage is provided that requires tightening of individual anchors, tightening shall be in accordance with procedures defined by the designer. Unless other specified, anchor tightening shall be performed: “ This change is considered editorial only and the TG voted unanimously to adopt the change (Affirmative: Krustulovic, Brannan, Hjorteset, and Ballard; Absent: Allen). Hoff and Hoptay will be contacted to withdraw their negative votes. Hanskat’s comment on the use of the word “tightening” when referring to anchors was addressed. The TG believes that the word “tightening” is appropriate when describing the minimum requirement for removing slack from the anchors. If the anchorage requires prestressing or tensioning, than a procedure is required, which is ultimately up to the engineer. This opinion was adopted by a TG unanimously vote (Affirmative: Krustulovic, Brannan, Hjorteset, and Ballard; Absent: Allen). Ballard’s editorial comment was considered accepted and the (a) and (b) requirements are reversed in their order to remove confusion that might arise as to the order of the minimum required anchor tightening and inspection procedure.



This change is considered editorial only and the TG voted unanimously to adopt the change (Affirmative: Krustulovic, Brannan, Hjorteset, and Ballard; Absent: Allen).

• Section R10.2.2 – Comment by Hoptay were addressed and found acceptable to the TG and the paragraph edited

accordingly. The reference to G16 will be clarified to specify that this document specifies application of statistical analysis to the corrosion data and that the reference to G46 will be clarified to specify that this document specifies the methods for examination and determination of size, shape, depth and density of pits, such as, visual inspection, non-destructive testing, radiographic examination, etc. This change is considered editorial only and the TG voted unanimously to adopt the change (Affirmative: Krustulovic, Brannan, Hjorteset, and Ballard; Absent: Allen).

• Section 10.3.X – Brannan’s suggested change to the text was included in this section. The TG believes that these changes will satisfy Hoptay’s negative vote. These changes will be reballoted during the Los Angeles convention. Pawski’s comments are considered persuasive by the TG and his suggested editorial changes are incorporated into the section. This change is considered editorial only and the TG voted unanimously to adopt the change (Affirmative: Krustulovic, Brannan, Hjorteset, and Ballard; Absent: Allen). At this point in the meeting, Allen joined the TG.

Chapter 5 – Load Factors • Table 5.1 – The TG considered Hanskat’s negative vote on reference to ACI 318 as persuasive. Reference to ACI 318 will

be replaced by ACI 350 throughout. This also addresses Brannan’s comment. The strength reduction factors in ACI 318 for strut-and-tie models will be defined the same as for shear and torsion. Hanskat will be contacted to withdraw his negative vote. Wu’s comment regarding developing separate tables showing the load factors for uplift (0.9 D) were considered, however, in the TG’s opinion, it is more efficient and customary to maintain the current format and require that .9 D be considered when uplift is important to the condition being evaluated, such as internal pressure. Hoptay and Hanskat’s negative vote on prestressing load factors was discussed. Hjorteset stated that prestress load factors are usually ignored except for design of the tendon anchorage region, where a load factor of 1.2 is typically considered. Otherwise prestress loads are considered on the capacity side of the equation. Hjosteset will provide additional information on this topic for consideration at the Los Angeles meeting. This will also address Wu’s comment. Ballard’s comments on load factors were discussed. He noted that the load factor for F of 1.4 is very conservative, considering that the fluid load is very carefully controlled and the possibility of overload was precluded by the freeboard limit on the tank. However, the fact that the hydrotest load factor is specified as 1.2 F and hydrotest is 1.25 times normal operating fluid pressure load, means that hydrotest will control in any event and therefore, the load factor of 1.4 on product fluid load is not relevant for LNG applications. Ballard also commented on the fact that the load factors for any accident load or earthquake (OBE and SSE) should be 1.0 and load factors for all loads in combination with OBE or SSE should also be 1.0. The two-level earthquake criteria, covers the probability of occurrence of the earthquake and the demands should be computed as the mean response of the system, therefore, no load factors should be included. This will be topic will be discussed further in Los Angeles.

• Table 5.2 – All comments received for Table 5.2 are the same as those for Table 5.1 and are therefore considered to be part of that discussion, except for Pawski’s comment.



Pawski’s comment on the wind directionality effect was discussed. The TG noted that the directionality of the wind is not relevant for the design of a vertical cylindrical tank since the design is axisymmetric. The other factor that will dictate the use of 1.3 is whether the site is outside hurricane region, per the ASCE 7 requirements and therefore, most LNG tanks will require a wind load factor of 1.6. This will be discussed further in Los Angeles. At this point in the meeting, Hjorteset had to leave for another engagement

Chapter 8 – Foundations • Table 8.1 and 8.2 – Allen presented his recommended safety factors for foundation design. Brannan and Ballard

disagreed with the use of the safety factors of 3.0 for normal operating conditions for foundations that are load tested during construction. Normally, this would permit a safety factor of 2.0. Also, Ballard stated that he believes that Eurocodes require a safety factor of 2.0 for OBE load combinations and 1.25 for SSE load combinations. Ballard will confirm the Eurocode safety factors for the Los Angeles meeting.



APPENDIX I: Record of withdrawn negative votes 10.2.x – Hoff withdrew his negative vote:

[email protected]

03/24/2008 08:03 AM

To [email protected]

cc

bcc

Subject Re: Reviewing your negative votes on 10.2.x

History: This message has been replied to.

Neven: The change is okay and I withdraw my negative. George George C. Hoff, P.E., DEng.PresidentHoff Consulting LLC250 Saddlewood LaneClinton, MS 39056Phone and Fax: 601-925-4070Email: [email protected]

Create a Home Theater Like the Pros. Watch the video on AOL Home.

Neven Krstulovic/U-Houston/ExxonMobil

03/20/2008 01:41 PM

To "George Hoff" <[email protected]>, [email protected]

cc

bcc

Subject Reviewing your negative votes on 10.2.x

George and Jo:

The TG has reviewed your comments and has changed text of the paragraph 10.2.x accordingly. The opinion of the TG was this change constitutes only an editorial change.

Please let me know if this change addresses your negative vote and, if so, would you like to withdraw your negative.

Thanks!

Yours,

Neven



APPENDIX II: Record of addressed comments and changes

ACI 376 / 376 R Last Update: 3/20/2008 Chapter V – Load Factors

Status after 3/18/2008 Ballot and March TG Houston Meeting Page 1 of 6

CHAPTER V – LOAD FACTORS

Approved Sections Section Approved with Comments Negative Vote

Table 5.1 – Primary Concrete Tank

Latest Text Reviewed Vote Committee Members’ COMMENTS Author RESPONSE Notes

I note the references to ACI 318. Should we also reference ACI 350 because we have adopted ACI 350 because ACI 376 is more appropriately covered by ACI 350?

Brannan General Comment on referencing ACI 318 instead of ACI 350

Ballot Results 2/17 – 3/18/08

Approved = 20 App. w. Com.= 1 Abst.= 6 Neg.= 1

(N) In general for liquid-containment shouldn't we be referencing ACI 350 instead of ACI 318? I assume this is ACI 318-05, not 318-08.

Hanskat

TG Houston Meeting – 3-18-08: Brannan, Hjorteset, Ballard, NKO All the load factors with references to ACI 318 are the same as the corresponding load factors in ACI 350. Therefore, all references to ACI 318 will be replaced with references to ACI 350. To contact Hanskat – regarding his negative. This is only an editorial change. Unanimously agreed by all the TG members present: Brannan, Hjorteset, Ballard, NKO.

In general we are referring to ACI-350 and not ACI-318. However, for loads it might be appropriate to refer to ASCE 7 “Minimum Design Loads for Buildings and Other Structures” in lieu of ACI-318 or ACI-350 as ACI load factors are based on ASCE 7 load factors.

Hjorteset General

Need to develop separate Tables showing the load factors applied for the uplift case, instead of combining with Table 5.1 & 5.2 (e.g. 0.9 for dead loads).

Wu TG Houston Meeting – 3-18-08: Brannan, Hjorteset, Ballard, NKO In the TG’s opinion it is more efficient to avoid multiple load tables, and to list both factors in the same table. Unanimously agreed by all the TG members present: Brannan, Hjorteset, Ballard, NKO.

On "prestressing loads", a) suggest that the load factor of 1.2 should be used (per ACI 318), instead of 1.15. b) The load factor "(1.0)" should be considered in the case of uplift case, and not for the case that the prestressing load is accurately defined.

Wu 2) Prestressing Loads – I to X Ballot Results 2/17 – 3/18/08


I am concerned about prestressing load factors listed in Table 5.1. ACI-318/350 and ASCE 7 do not list load factors for prestressing loads. For bridge design load factors for prestressing loads are sometimes given, however, the load factors used are only for “Forces and moments transferred from members containing post-tensioning steel to other members upon application of the post-tensioning force” or for secondary effects. (For primary post-tensioning effects, load factors have no meaning). For bridge design, the load factors for secondary effects are typically 1.0 for all load combinations. The 1.2 load factor referenced to ACI 318 paragraph 9.2.5 is for local anchorage zone design and not for global design. A load factor of 1.2 on maximum initial stressing load for post-tensioned anchor design should be

Hjorteset

TG Houston Meeting – 3-18-08: Brannan, Hjorteset, Ballard, NKO At the beginning of this chapter define that load factors specified in this chapter should be used in conjunction with strength reduction factors defined in ACI 350. For Strut-and-Tie models use 0.75, as per ACI 318. Prestressing Loads say: A load factor of 1.0 should be used for all prestressing loads except in the case of the anchorage zones. For post-tensioned anchorage zone design, a load factor of 1.2 shall be applied to the maximum prestressing steel jacking force, as per ACI 350. Commentary: The load factor of 1.2 applied to the maximum tendon jacking force results in a




maintained because a single tendon may by accident be severely overstressed. However, a load factor of 1.2 for global effects might be high (conservative) when the load factors are applied to load effects from initial stressing loads or even final stressing loads (including seating loss). It would be prudent to look further into the magnitude of load factors for post-tensioning and what post-tensioning loads should be used as an upper and lower bound. For example, upper bound might be initial post-tensioning including tendon friction, anchor seating and elastic shortening of concrete. While a lower bound includes losses due to creep, shrinkage, and relaxation. (N) In 2) it states: "1.15 - usually used for PSC". "Usually" is not mandatory language..

Hanskat

Proposed 1.15 factor for prestressing loads other than the anchorage zone. Before any of the proposed load factors can be approved, the strength reduction factors need to be defined and approved. Both factors need to form a consistent philosophy of design.

Hoptay

design load of about 113 percent of the specified prestressing steel yield strength, but not more than 96 percent of the nominal ultimate strength of the prestressing steel. This compares well with the maximum attainable jacking force, which is limited by the anchor efficiency factor. Unanimously agreed by all the TG members present: Brannan, Hjorteset, Ballard, NKO. Hjorteset to further pursue the issue and contact Hoptay

7) Construction – Operation Loads Ballot Results 2/17 – 3/18/08


(E) In item 7) change CONSTRUCTION-OPERATION to CONSTRUCTION AND OPERATION

Hanskat Change introduced.

15 ) Environmental Loads Ballot Results 2/17 – 3/18/08


• In section 15 on Wind there is no factor under any of the abnormal loading conditions but there is 1.0 in the notes. This is inconsistent with the presentations in the other sections. Please clarify.

• One of the notes in Section 19 refers to the tank's response if it is on piles and seems to

imply that a different response will occur if the tank is on a shallow foundation. Please clarify.

Allen TG Houston Meeting – 3-18-08: Brannan, Allen, Ballard, NKO Factor of 1.0 removed from the commentary To address the second comment, introduce the following change: “Explosion and impact loads generally have little of no effect on the primary tank of a double wall tank. However, if the tank is on piles the response of the entire structure may induce forces in the primary container.” This is only an editorial change. Unanimously agreed by all the TG members present: Brannan, Allen, Ballard, NKO.



Ballard, Thomas A Load factors for product pressure are high considering that the probability of the fill level exceeding the maximum design level is limited by the freeboard and that reliability of the design is pretty high.

However, if the load factor of 1.4(D + F) is the only load combination where 1.4F is applied, then this should not control the design since a 1.2 factor on hydrotest (1.25*F) will control the design. Load factors for dead loads when combined with Spill and Spill+SSEalt should be 1.0. The table implies that dead load is not combined with spill or spill+SSEalt, which is not the case. Load factors for OBE or SSE should be 1.0 with different performance limits set for OBE and SSE. For a two-level earthquake design, the probability of failure is implicit in the earthquake return period and all other factors should be based on best estimate. If additional factors are introduced for probability that the loads exceed their best estimated value or probability that the design does not perform as expected, then the probability of failure is skewed to the more conservative side.

Godejord, Arnstein See pdf file) 2 – IV: Load factor less than 1.0 should be considered. Less prestressing than intended may lead to leakage 2 – VII: Load factor other than 1.0 seems unreasonable 3 – VII: Load factor other than 1.0 seems unreasonable 17 – VII: The load factor for OBE should be 1.0. Load factor of 1.3 is to strict.



Table 5.2 – Secondary Concrete Tank

Latest Text Reviewed Vote Committee Members’ COMMENTS Author RESPONSE Notes I note the references to ACI 318. Should we also reference ACI 350 because we have adopted ACI 350 because ACI 376 is more appropriately covered by ACI 350?

Brannan General Comment on referencing ACI 318 instead of ACI 350



(N) In general for liquid-containment shouldn't we be referencing ACI 350 instead of ACI 318? I assume this is ACI 318-05, not 318-08.

Hanskat

TG Houston Meeting – 3-18-08: Brannan, Hjorteset, Ballard, NKO All the load factors with references to ACI 318 are the same as the corresponding load factors in ACI 350. Therefore, all references to ACI 318 will be replaced with references to ACI 350. To contact Hanskat – withdrawing his negative? This is only an editorial change. Unanimously agreed by all the TG members present: Brannan, Hjorteset, Ballard, NKO.

In general we are referring to ACI-350 and not ACI-318. However, for loads it might be appropriate to refer to ASCE 7 “Minimum Design Loads for Buildings and Other Structures” in lieu of ACI-318 or ACI-350 as ACI load factors are based on ASCE 7 load factors.

Hjorteset General

Need to develop separate Tables showing the load factors applied for the uplift case, instead of combining with Table 5.1 & 5.2 (e.g. 0.9 for dead loads).

Wu TG Houston Meeting – 3-18-08: Brannan, Hjorteset, Ballard, NKO In the TG’s opinion it is more efficient to avoid multiple load tables, and to list both factors in the same table. Unanimously agreed by all the TG members present: Brannan, Hjorteset, Ballard, NKO.

On "prestressing loads", a) suggest that the load factor of 1.2 should be used (per ACI 318), instead of 1.15. b) The load factor "(1.0)" should be considered in the case of uplift case, and not for the case that the prestressing load is accurately defined.

Wu E-2) Prestressing Loads – I to X Ballot Results 2/17 – 3/18/08


I am concerned about prestressing load factors listed in Table 5.1. ACI-318/350 and ASCE 7 do not list load factors for prestressing loads. For bridge design load factors for prestressing loads are sometimes given, however, the load factors used are only for “Forces and moments transferred from members containing post-tensioning steel to other members upon application of the post-tensioning force” or for secondary effects. (For primary post-tensioning effects, load factors have no meaning). For bridge design, the load factors for secondary effects are typically 1.0 for all load combinations. The 1.2 load factor referenced to ACI 318 paragraph 9.2.5 is for local anchorage zone design and not for global design. A load factor of 1.2 on maximum initial stressing load for post-tensioned anchor design should be maintained because a single tendon may by accident be severely overstressed. However, a load factor of 1.2 for global effects might be high (conservative) when the load factors are applied to load effects from initial stressing loads or even final stressing loads (including seating loss). It would be prudent to look further into the magnitude of load factors for post-tensioning and what post-tensioning loads should be used as an upper and lower bound. For example, upper bound might be initial post-tensioning including tendon friction, anchor seating and elastic shortening of concrete. While a lower bound includes losses due to creep, shrinkage, and relaxation.

Hjorteset

TG Houston Meeting – 3-18-08: Brannan, Hjorteset, Ballard, NKO At the beginning of this chapter define that load factors specified in this chapter should be used in conjunction with strength reduction factors defined in ACI 350. For Strut-and-Tie models use 0.75, as per ACI 318. Prestressing Loads say: A load factor of 1.0 should be used for all prestressing loads except in the case of the anchorage zones. For post-tensioned anchorage zone design, a load factor of 1.2 shall be applied to the maximum prestressing steel jacking force, as per ACI 350. Commentary: The load factor of 1.2 applied to the maximum tendon jacking force results in a design load of about 113 percent of the specified prestressing steel yield strength, but not more than 96 percent of the nominal ultimate strength of the prestressing steel. This compares well with the maximum attainable jacking force, which is limited by the anchor efficiency factor. Unanimously agreed by all the TG members present: Brannan, Hjorteset, Ballard, NKO.



Latest Text Reviewed Vote Committee Members’ COMMENTS Author RESPONSE Notes (N) In 2) it states: "1.15 - usually used for PSC". "Usually" is not mandatory language..

Hanskat

The proposed load factor reduction from 1.2 to 1.0 for the OBE load is not consistent with the NFPA 59A (B.2)statement that the OBE is designed in accordance with conventional engineering procedures and criteria. Also it is not consistent to have the same load factor as the SSE event which is the limit state check.

Hoptay

Hjorteset to further pursue the issue and contact Hoptay

E-7) Construction – Operation Loads Ballot Results 2/17 – 3/18/08


(E) In item 7) change CONSTRUCTION-OPERATION to CONSTRUCTION AND OPERATION

Hanskat Change introduced.

E-15) Environmental Loads - Wind E-15 environmental wind loads are 1.6 without directionality factor and 1.3 with directionality factor per ACI 318. This is correct, but ASCE 7 commentary says that there is another criterion to satisfy in order to use 1.3, and that is the site has to be outside hurricane region. For most import facilities this would require the 1.6 factor and not the 1.3.

Pawski TG Houston Meeting – 3-18-08: Brannan, Allen, Ballard, NKO In the TG’s opinion ACI 350 should be followed. Ballard to review the issue further and update the Editorial TG accordingly. Unanimously agreed by all the TG members present: Brannan, Allen, Ballard, NKO.

E-18) Seismic Loads - SSE E-19) Explosion and Impact E-20) Fire



Agree with all proposed load factors with the exception of "abnormal" combination 18 - SSE; 19 Explosion & Impact; 20 Fire where a load factor of 1.0 is specified. A load factor of 1.05 would be recommendable. See EN 14620-3 Table 1 and BS 7777 Part 3 Table A.3 and given the load factors proposed for the inner tank in Table 5.1. Alternatively, but less transparent and satisfactory, values may be specified for flux, explosion etc.

Douglas


Status after 3/18/2008 Ballot and March TG Houston Meeting Page 6 of 6 Ballard, Thomas A

Load factors for product pressure are high considering that the probability of the fill level exceeding the maximum design level is limited by the freeboard and that reliability of the design is pretty high. However, if the load factor of 1.4(D + F) is the only load combination where 1.4F is applied, then this should not control the design since a 1.2 factor on hydrotest (1.25*F) will control the design. Load factors for dead loads when combined with Spill and Spill+SSEalt should be 1.0. The table implies that dead load is not combined with spill or spill+SSEalt, which is not the case. Load factors for OBE or SSE should be 1.0 with different performance limits set for OBE and SSE. For a two-level earthquake design, the probability of failure is implicit in the earthquake return period and all other factors should be based on best estimate. If additional factors are introduced for probability that the loads exceed their best estimated value or probability that the design does not perform as expected, then the probability of failure is skewed to the more conservative side.

Godejord, Arnstein

Load factor less than 1.0 for prestressing loads should be considered Load factor for earthquake OBE should be set to 1.0

TABLE 5.1 - LOAD FACTORS FOR THE ULS OF THE PRIMARY CONTAINER

Operation

EARTHQUAKE

IV -

OPE

RAT

ION

V - S

PILL

VI -

SPIL

L +

SSE

aft

NA NA

1.15(1.0)

LOADING CONDITION NORMAL LOADING CONDITION ABNORMAL LOADING CONDITION

1.2 (0.9)

1.2 (0.9)

1.2/1.4 (0.9)

LOADING TYPES

I - C

ON

STR

UC

TIO

N

II - I

NST

ALLA

TIO

N

III -

TEST

ING

AN

D

CO

MM

ISSI

ON

ING

IX -

EXPL

OSI

ON

AN

D

IMPA

CT

X - F

IRE

VII -

OBE

VIII

- SSE

1 (0.9)

1 (0.9)

Construction Loads 1.2 - dead load combined with other loads - as per ACI 350 Eq. 9-2 to 9-5(0.9) - when dead load combined with other loads and is beneficial - as per ACI 350 Eq. 9-6 and Eq.9-7.Installation Loads

Testing and Commissioning

1.2 (0.9)

1 (0.9)1) DEAD LOADS 1.2

(0.9)

1.15 / 1.2*(1.0)

1.15 / 1.2*(1.0)

Spill + SSE aft

SSE

OBE

SSE

SpillSpill + SSE aft

---

1.15(1.0) 1.00

OBE

NA

1.2

NA NA2) PRESTRESSING LOADS 1.15 / 1.2*

(1.0)

1.2 - product pressure combined with other loads - as per ACI 350 Eq. 9-2

1.00 1.00

1.00 1.00

NA

1.15 - usually used for PSC - ***CITE SOURCE**1.2* - for post-rensioning anchorage zones a load factor of 1.2 shall be applied to the maximum prestressing steel jacking force, as per ACI 350paragraphs 9.2.5 and R9.2.5(1.0) - the prestressing load is accurately defined during prestressing operations.

1.2 1.00---

----

1.2

----

Fire


1.0 - for an extreme event

1.2 / 1.4

Explosion and ImpactFire

1.00 ---- NA 1.2 1.00

NA NA ----

1.2 NA

----

OBE

---- ----

6) FIRE ---- ---- ---- ---- NA

Installation LoadsTesting and Commissioning

NA

THERMAL AND/OR MOISTURE

4) NORMAL 1.2 1.2

5) SPILL ---- ----

---- 1.00

Construction Loads

---- ----

SpillNASpill + SSE alt

SSE

9) TANK TESTING LOADS

----

Explosion and Impact

1.6 ----

8) INSTALLATION LOADS

---- 1.2 / 1.4(0.9)

CO

NST

RU

CTI

ON

AN

D

CO

MM

ISIO

NIN

G L

OA

DS ---- ---- NA ---- ----NA ---- ----

---- ---- NA

----

1.2 ---- NA ----

---- NA

----

---- ---- NA ----

NA ---- ---- ----

---- 1.2 ----

7) CONSTRUCTION and OPERATION

NA ---- ---- 10) THERMALLY-INDUCED TANK COOLING AND FILLING LOADS

----

1.2 - thermal / mositure combined with other loads - as per ACI 350 Eq. 9-2

NA NA

Fire



Construction Loads Installation Loads

Testing and Commissioning Operation

1.15 - usually used for PSC - ***CITE SOURCE**1.2* - for post-rensioning anchorage zones a load factor of 1.2 shall be applied to th emaximum prestressing steel jacking force, as per ACI 350paragraphs 9.2.5 and R9.2.5(1.0) - the prestressing load is accurately defined during prestressing operations.


1.2 - product pressure combined with other loads - consistent with ACI 350 Eq. 9-21.4 - product pressure combined only with the dead load - consistent with ACI 350 Eq. 9-1NOTE: during th etesting and comissioning phase, th e"product" denotes liquid used in hydrostatic testing

Construction Loads Installation Loads

Testing and Commissioning Operation

3) PRODUCT PRESSURE

Operation

FireExplosion and Impact

1.2 - dead load combined with other loads, including seismic laod - as per ACI 350 Eq. 9-2 to 9-5(0.9) - when dead load combined with other loads and is beneficial - as per ACI 350 Eq. Eq.9-7.

1.2 - dead load combined with other loads - as per ACI 350 Eq. 9-2 to 9-51.4 - when dead load is combined only with the product pressure - as per ACI 350 Eq. 9-1(0.9) - when dead load combined with other loads and is beneficial - as per ACI 350 Eq. 9-6 and Eq.9-7.

Spill NA

OBE

SSE


EARTHQUAKE

IV -

OPE

RAT

ION

V - S

PILL

VI -

SPIL

L +

SSE

aft


LOADING TYPES

I - C

ON

STR

UC

TIO

N

II - I

NST

ALLA

TIO

N

III -

TEST

ING

AN

D

CO

MM

ISSI

ON

ING

IX -

EXPL

OSI

ON

AN

D

IMPA

CT

X - F

IRE

VII -

OBE

VIII

- SSE

Operation

Operation

Operation

OBE

SpillSpill + SSE alt

Testing and commissioning loads are controlled and monitored and as such are treated as dead loads for a proposed load factor of 1.2 per ACI 350 Equation 9-2 since they are combined with thermal loads 1.2 - dead load combined with other loads - as per ACI 350 Eq. 9-2 to 9-5(1.0) - Testing and commissioning loads are controlled and monitored, and are hence accurately defined during prestressing operations.

The loading types are Construction and Commissioning therefore these loadings are not applicable to in-service conditions.

NA


Construction Loads 1.6 - Construction loads are catagorized as live loads - as per ACI 350 Eq. 9-2.

Installation Loads

Installation loads are better defined and less variable than construction loads and are treated as dead loads per ACI 350:1.2 - dead load combined with other loads - as per ACI 350 Eq. 9-2 to 9-51.4 - when dead load is combined only with the product pressure - as per ACI 350 Eq. 9-1(0.9) - when dead load is combined with other loads and is beneficial - as per ACI 350 Eq. 9-6 and Eq.9-7.

Fire

1.2 1.2 NA

The loading types are Construction and Commissioning therefore these loadings are not applicable to in-service conditions.


SSE

11) SHRINKAGE 1.2 1.2 1.00 1.00NA 1.2 1.00

SSE

Construction Loads

1.2 - shrinkage combined with other loads - as per ACI 350 Eq. 9-2Installation Loads


OBE

SpillNASpill + SSE aft

1.2 - shrinkage combined with other loads - as per ACI 350 Eq. 9-2

1.00 1.001.2 1.001.2

Construction LoadsInstallation Loads



NA



Fire



NA NA12) CREEP 1.2


1.2 1.2


SSE

OBE


EARTHQUAKE

IV -

OPE

RAT

ION

V - S

PILL

VI -

SPIL

L +

SSE

aft


LOADING TYPES

I - C

ON

STR

UC

TIO

N

II - I

NST

ALLA

TIO

N

III -

TEST

ING

AN

D

CO

MM

ISSI

ON

ING

IX -

EXPL

OSI

ON

AN

D

IMPA

CT

X - F

IRE

VII -

OBE

VIII

- SSE

Operation

Operation

Operation

Operation

Explosion and impact loads generally have little of no effect on the primary tank of a double wall tank. However, the response of the entire structure may induce forces in the primary container.

1.00 ----

NA


13) GENERAL LIVE LOADS



Load factor per ACI 349 Section 9.2.1, Eq.10

1.0 - as per ACI 350 Eq. 9-5 and Eq.9-7 and ACI 349 Section 9.2.1 Eq.4


---- ----

SSE

OBE

OBE

SSE

1.6 / 0.5** 1.6 / 0.5** 1.6 / 0.5** 1.6 / 0.5**1.6 / 0.5** NA NA 1.00

Fire

NA


OBE

OBE

SSE

Construction Loads

1.6 - as per ACI 350 Eq. 9-3.0.5 - can be reduced to 0.5 when wind effects consideredd - as per ACI 350 Eq. 9.4

1.6 - general live loads (e.g., piping loads) - as per ACI 350 Eq. 9.20.5** - can be reduced to 0.5 where live load is greater than 100 psf - as per ACI 350 paragraph 9.2.1 (a)

NA

NA NA 1.2


14) DIFFERENTIAL SETTLEMENT 1.2 1.2

1.2 - differential settlement combined with other loads - as per ACI 350 Eq. 9-2


NA




Fire

1.0 - as per ACI 349 which a load factor of 1.0 for live loads such as piping loads when combined with SSE seismic loads.

NA


1.001.001.2 1.2 1.00

1.0 - for an extreme eventExplosion and ImpactFire

15) ENVIRONMENTAL LOADS: Wind 1.6 / 1.3 N.A. N.A.

SSE

NA

1.3 - to be applied when a directionality factor has not been used, as per ACI 350 Eq. 9.2 and paragraph 9.2.1 (b)1.6 - to be applied when a directionality factor has been used, as per ACI 350 Eq. 9.2 and paragraph 9.2.1 (b)Wind load generally has little of no effect on the primary tank of a double wall tank. However if piping or other connection do transfer load to theprimary container the listed load factors should be used accordingly.

---- ---- N.A. NA NA ----




OBE

SSE

16) ENVIRONMENTAL LOADS: Other ---- ---- 1.6 / 0.5 N.A. N.A. N.A.

---- ---- ---- NA NA 1.3 ----

NA NA

SEISMIC LOADS17) OBE

18) SSE



Construction Loads

NA ----

---- ----

NA

----

NASpill + SSE aft

1.0 NANA NA ---- ---- ----

---- NA NA19) EXPLOSION AND IMPACT ---- ---- ---- ---- ---- 1.0 ----


EARTHQUAKE

IV -

OPE

RAT

ION

V - S

PILL

VI -

SPIL

L +

SSE

aft


LOADING TYPES

I - C

ON

STR

UC

TIO

N

II - I

NST

ALLA

TIO

N

III -

TEST

ING

AN

D

CO

MM

ISSI

ON

ING

IX -

EXPL

OSI

ON

AN

D

IMPA

CT

X - F

IRE

VII -

OBE

VIII

- SSE

a) The short term and long term effects should be considered for cases of shrinkage, creep, differential settlement, and prestressing tendon relaxation.b) Per ACI 350, Section 9.2.6, the environmental durability factor, Sd, should be considered in the concrete design.

---- ---- 1.00 ---- ---- NA NA20) FIRE ---- ---- ----

TABLE 5.2 - LOAD FACTORS FOR THE ULS OF THE SECONDARY CONTAINER


1.0 - for an extreme event0.0 - should be used for vacuum loading since due to the generation of vapor during a spill event vacuum loading is nota credible event

E3-a) PRODUCT PRESSURE: Vapor / Gas / Vacuum

Spill

Spill

Operation

Spill + SSE aft

Fire1.0 - for an extreme eventNote that for the abnormal fire-loafing load combination, it is assumed that the roof has been lost and hence there is

no gas/vapor pressure.

OBE

Operation

Operation - OBE No Spill Primary Container Full of Empty

SSE


OBE

Spill

1.00

SSE


1.2 - product pressure combined with other loads - consistent with ACI 350 Eq. 9-21.4 - product pressure combined only with the dead load - consistent with ACI 350 Eq. 9-1For the secondary container product gas/vapor pressure develops first during commissioning. The vapor pressure isconsidered to be the same as a fluid pressure, as defined in ACI 350.

Installation Loads

----

Construction Loads

1.0 - for an extreme eventFor the secondary container hydrostatic product pressure is not included in any load case that does not include a spill.


1.2 - product pressure combined with other loads - consistent with ACI 350 Eq. 9-2For the secondary container product pressure is not included in any load case that does not include a spill.

1.00 1.00 ----1.2 / 1.4 1.0 / 0.0+ 1.0 / 0.0+

1.15 - usually used for PSC - ***CITE SOURCE**1.2* - for post-rensioning anchorage zones a load factor of 1.2 shall be applied to th emaximum prestressing steeljacking force, as per ACI 350 paragraphs 9.2.5 and R9.2.5(1.0) - the prestressing load is accurately defined during prestressing operations.

Installation Loads


Proposed change from 1.2 (1.0)The proposed 1.2 Load Factor is consistent with ACI 350 Eq. 9-2 since it is assumed that the dead load will becombined with other operating loads such as settlement and live load. The (0.9) consistent with Eq. 9-6 and Eq.9-7when the higher dead load reduces the effect of other loads.

Construction Loads

1.0 - for an extreme event(0.9) - when dead load combined with other loads and is beneficial - as per ACI 350 Eq. 9-6 and Eq.9-7.Spill + SSE aft

SSE

1.15 / 1.2*(1.0)

1.15 / 1.2*(1.0)

1.15(1.0)

1.0 (0.9)

1.00 (0.9)

1.15(1.0)

Fire

1.15 - usually used for PSC - ***CITE SOURCE**1.2* - for post-rensioning anchorage zones a load factor of 1.2 shall be applied to the maximum prestressing steeljacking force, as per ACI 350 paragraphs 9.2.5 and R9.2.5(1.0) - the prestressing load is accurately defined during prestressing operations.1.0 - for an extreme event

Fire

---- ----

OBE

1.2

1 (0.9)

1.00 1.00

1.0 - for an extreme event(0.9) - when dead load combined with other loads and is beneficial - as per ACI 350 Eq. 9-6 and Eq.9-7.

1.0 - for an extreme event(0.9) - when dead load combined with other loads and is beneficial - as per ACI 350 Eq. 9-6 and Eq.9-7.1.2 - dead load combined with other loads, including seismic laod - as per ACI 350 Eq. 9-2 to 9-5(0.9) - when dead load combined with other loads and is beneficial - as per ACI 350 Eq. Eq.9-7.

1.2 - dead load combined with other loads - as per ACI 350 Eq. 9-2 to 9-5(0.9) - when dead load combined with other loads and is beneficial - as per ACI 350 Eq. 9-6 and Eq.9-7.

1.00

1.2 - dead load combined with other loads - as per ACI 350 Eq. 9-2 to 9-51.4 - when dead load is combined only with the product pressure - as per ACI 350 Eq. 9-1(0.9) - when dead load combined with other loads and is beneficial - as per ACI 350 Eq. 9-6 and Eq.9-7.

1.2 (0.9)

1.2 (0.9)

1.0 (0.9)

1.0 (0.9)

1.2 (0.9)

1 (0.9)

1.2 (0.9)

1.2/1.4 (0.9)

1 (0.9)

IX -

EX

PLO

SIO

N A

ND

IM

PA

CT

X - F

IRE

EARTHQUAKE

E-2) PRESTRESSING LOADS

NORMAL LOADING CONDITIONLOADING CONDITION ABNORMAL LOADING CONDITION

1.15 / 1.2*

(1.0)

E-1) DEAD LOADS

VIII

- SS

E

I - C

ON

STR

UC

TIO

N

II - I

NS

TALL

ATI

ON

III -

TES

TIN

G A

ND

C

OM

MIS

SIO

NIN

G

IV -

OP

ER

ATI

ON

V - S

PIL

L

VI -

SP

ILL

+ S

SE

aft

VII -

OB

E


---- ---- ---- ---- 1.00

---- ---- 1.2

1.00

LOADING TYPES



Spill + SSE aft

E3-b) PRODUCT PRESSURE: Liquid



IX -

EX

PLO

SIO

N A

ND

IM

PA

CT

X - F

IRE

EARTHQUAKENORMAL LOADING CONDITIONLOADING CONDITION ABNORMAL LOADING CONDITION

VIII

- SS

E

I - C

ON

STR

UC

TIO

N

II - I

NS

TALL

ATI

ON

III -

TES

TIN

G A

ND

C

OM

MIS

SIO

NIN

G

IV -

OP

ER

ATI

ON

V - S

PIL

L

VI -

SP

ILL

+ S

SE

aft

VII -

OB

E

LOADING TYPES



Testing and commissioning loads are controlled and monitored and as such are treated as dead loads for a proposed load factor of 1.2 per ACI 350 Equation 9-2 since they are combined with thermal loads 1.2 - dead load combined with other loads - as per ACI 350 Eq. 9-2 to 9-5(1.0) - Testing and commissioning loads are controlled and monitored, and are hence accurately defined during prestressing operations.

1.2


Installation loads are better defined and less variable than construction loads and are treated as dead loads per ACI 350:1.2 - dead load combined with other loads - as per ACI 350 Eq. 9-2 to 9-51.4 - when dead load is combined only with the product pressure - as per ACI 350 Eq. 9-1(0.9) - when dead load is combined with other loads and is beneficial - as per ACI 350 Eq. 9-6 and Eq.9-7.

---- ----


1.0

Fire

The loading types are Construction and Commissioning therefore these loadings are not Applicable to in-service condit

Spill + SSE aft

Fire

Spill

Construction Loads

1.2 - shrinkage combined with other loads - as per ACI 350 Eq. 9-2Installation LoadsTesting and Commissioning

Operation

E-12) CREEP

----

---- ---- ----


---- ---- ----

Operation

----

----

Spill

----

----

1.001.00 1.00 1.2 1.00


1.01.0 1.0 1.2 1.0

1.001.2

1.2

1.00 ----

---- ---- ----

1.2

---- 1.001.00 ---- ---- ---- ----

1.6 ---- ----


----


-------- --------


---- ---- ---- ---- ----

----


---- ---- ----



1.2

---- ----

1.2 1.2

---- 1.2

----

----

1.6 - Construction loads are catagorized as live loads - as per ACI 350 Eq. 9-2.

----

1.2 1.2

----

---- 1.2 / 1.4(0.9) ----

---- ---- 1.2

THERMAL AND/OR MOISTURE

E-4) NORMAL

E-6) FIRE

Spill + SSE aft


E-5) SPILL




Testing and Commissioning 1.2 - shrinkage combined with other loads - as per ACI 350 Eq. 9-2

E-8) INSTALLATION

E-9) TANK TESTING LOADS

E-10) THERMALLYINDUCED TANK COOLING AND

FILLING LOADS

CO

NST

RU

CTI

ON

AN

D

CO

MM

ISSI

ON

ING

LO

AD

S

E-11) SHRINKAGE

OBE

Fire

OperationSpill

Construction Loads


Installation Loads


OperationSpill

Spill + SSE aft

Spill + SSE aft

Fire

OBE

OBE


1.2 1.2

1.00 ---- ---- ---- ----

---- ---- 1.001.2 1.2

E-7) CONSTRUCTION and OPERATION

SSE

SSE

SSE

SSE

OBE


IX -

EX

PLO

SIO

N A

ND

IM

PA

CT

X - F

IRE

EARTHQUAKENORMAL LOADING CONDITIONLOADING CONDITION ABNORMAL LOADING CONDITION

VIII

- SS

E

I - C

ON

STR

UC

TIO

N

II - I

NS

TALL

ATI

ON

III -

TES

TIN

G A

ND

C

OM

MIS

SIO

NIN

G

IV -

OP

ER

ATI

ON

V - S

PIL

L

VI -

SP

ILL

+ S

SE

aft

VII -

OB

E

LOADING TYPES

a) The short term and long term effects should be considered for cases of shrinkage, creep, differential settlement, and prestressing tendon relaxation.b) Per ACI 350, Section 9.2.6, the environmental durability factor, Sd, should be considered in the concrete design.

--- 1.0 ---- ---- ---- ---- ---- E-20) FIRE ---- ---- ----

Explosion and Impact 1.0 - for an extreme event

Spill + SSE aft

---- ---- ---- ---- ----

The proposed Load factor per ACI 350 Eq. 9-5 and Eq.9-7 and ACI 349 Section 9.2.1.4

----

These loads are not required to be combined with abnormal loadings

Spill + SSE aft


OBE


OBE

Spill


Operation

Construction Loads

Spill + SSE aft

Construction Loads

1.2 - differential settlement combined with other loads - as per ACI 350 Eq. 9-2Installation LoadsTesting and Commissioning

OperationSpill

OperationSpill

Spill + SSE aft



E-15) ENVIRONMENTAL LOADS: Wind

Testing and CommissioningOperation


E-13) GENERAL LIVE LOADS ---- ----

Fire 1.0 - for an extreme event

----

Wind loads are not required to be combined with abnormal loadings


---- 1.6 / 1.3 1.6 / 1.3

1.2 1.2 1.2

----


---- 1.0

1.3 ----

----

General live loads are not required to be combined with abnormal loadings

1.6 - as per ACI 350 Eq. 9-3.0.5 - can be reduced to 0.5 when wind effects consideredd - as per ACI 350 Eq. 9.4


1.0

1.6 / 1.3 1.6 / 1.3 ---- ---- ----

---- ----

---- ----

----

----

----

1.0

---- ---- 1.6 / 0.5** 1.6 / 0.5** 1.6 / 0.5** ----

1.0 1.0

----

1.01.0 1.2


1.6 / 0.5**

1.2

---- ----

1.3 - to be applied when a directionality factor has not been used, as per ACI 350 Eq. 9.2 and paragraph 9.2.1 (b)1.6 - to be applied when a directionality factor has been used, as per ACI 350 Eq. 9.2 and paragraph 9.2.1 (b)

1.6 / 0.5 1.6 / 0.5 1.6 / 0.5 1.6 / 0.5 ---- ---- ----

---- ---- ----

Load factor per ACI 349 Section 9.2.1, Eq.10

1.0 - as per ACI 350 Eq. 9-5 and Eq.9-7 and ACI 349 Section 9.2.1 Eq.4

1.0

---- ---- ---- ----

---- ---- ---- ----SEISMIC LOADS

E-17) OBE

E-18) SSE

E-19) EXPLOSION AND IMPACT

SSE

Spill + SSE aft



E-16) ENVIRONMENTAL LOADS: Other

E-14) DIFFERENTIAL SETTLEMENT

OBE

Spill

Fire

OBE

OBE

SSE

SSE

SSE

SSE

ACI 376 / 376 R Last Update: 3/20/2008 Chapter VI – Analysis and Design

Status after 2/19 to 3/20 Ballot and TG Houston Meeting Page 1 of 2

CHAPTER 6 – ANALYSIS AND DESIGN

Approved Sections


"Constitutive Model" should be added to the definition list. R6.1.1 states that constitutive models should be in accordance with European Code EC2. Since EC2 is not a readily available document for most engineers it is suggested that the specific requirements of EC2 be defined in the Chapter 1.

Hoptay A) the existing text of paragraph 6.1.1 be replaced with:

6.1.1 – Required analysis - The containment structure shall be analyzed as an integrated structure that includes the foundation, wall, roof, contained liquid, liner or portion of the liner that is assumed to act compositely with the concrete structure. The effects of discontinuities shall be considered. Constitutive models, assumed values and details used in the analysis shall be approved by the owner/engineer.

B) the following text be added at the end of the commentary paragraph R6.1.1:

… The effect of soil stiffness should be included in the analysis as defined in 6.1.2. For load conditions 3.1.15 and 3.1.16, which include severe thermal loading conditions, the structure should be analyzed for the entire transient history up to and including steady state. Both maximum and minimum design ambient temperatures should be used as the initial temperature profiles for the analysis of all loading conditions. For load condition 3.1.15, the temperature for the entire structure, including the roof, should be estimated. The temperature resulting from vapor generation during roll-over and spill events shall be taken into account. The structural model for load conditions 3.1.15 and 3.1.16 should consider the entire temperature time history and be analyzed on the basis of transient inelastic response. Serviceability requirements should be checked both during the transient and steady state temperature profiles. The analysis for these thermal load conditions should take into account the effect of cracking and tension stiffening. Cracking and tension stiffening should be included by appropriate modification of the material stress strain relationship or by the use of finite elements that have the capability of cracking under tension, and crushing under compression as well as the ability to include reinforcing steel.

Ballot 2/19/08 – 3/20/08


It appears that the characteristics demanded of an FEM-Program to fulfill the requiremets of this clause are such that only complex university or research type programs will suit.This may lead to a dangerous situation in that the results are difficult to verify = "black box" syndrome. The use of less complex programs, which, with experience supply sensible and verifiable results, may be considered inapplicable as they would fail to meet a stringent specification contained here. This aspect should be reconsidered.

Douglas

TG Houston Meeting – 3-18-08: Brannan, Hjorteset, Ballard, NKO The concrete model cited by EC2 is reasonably widely used and available. The model is a part of various standard F.E. packages such as Adina, ABAQUS, etc. Furthermore, EC2 is believed to be easier to access and obtain than the original papers on which the code model is based. Therefore, the reference to EC2 is maintained. However, the TG advises that the original model references be included in the commentary section where EC2 is referenced. Unanimously agreed by all the TG members present: Brannan, Hjorteset, Ballard, NKO.

Include original references for the model used by the EC2 code. Also include full reference to EC2, including listing it in the reference section.

It should be reviewed whether non-linear methods for the outer concrete tank should be limited to the SSE case. Our experience shows that this makes little sense for the OBE+Operation case.

Douglas R6.1.2.1 - The ballot proposes that Mash’s negative vote on the ballot from 9/13/2007 to 10/13/2007 is found accepted in principle in part. To address raised issues, the following text is proposed to replace the existing text:

R6.1.2.1 – Both linear and non-linear analysis may be used in determining seismic forces. In general, linear analysis is used in the case of low seismic regions, while non-linear analysis is used in regions with higher seismicity. When seismic forces are determined using a linear elastic approach, the

Ballot 2/19/08 – 3/20/08

Approved = 21 App. w. Com.= 3 Abst.= 4

Include the terms "demonstrated explicitly" Remove "static", as nonlinear is by definition not static

Mash

TG Houston Meeting – 3-18-08: Brannan, Hjorteset, Ballard, NKO • Mash Comment 1: term “demonstrated explicitly" included • Mash Comment 2: static analysis can be linear or non-linear,

depending on the range of material behavior. The following wording was added (e.g., “push-over”).

• Wu’s comment: included R6.1.2.1 – Both linear and non-linear analysis may be used in determining seismic forces. In general, linear analysis is used in the case of low seismic regions and/or OBE case, while non-linear

Ballard: I think the discussion of ASCE 4 should be expanded a bit. ASCE 4 discusses selection of material properties, damping, discretization of mass, etc. This

ACI 376 / 376 R Last Update: 3/20/2008 Chapter VI – Analysis and Design



response modification factor should be taken as R=1. A force reduction factor of R > 1 may be used if it can be shown, by means of dynamic or static nonlinear analyses, that the structure meets or exceeds the performance criteria prescribed in this Code. Guidance for selecting material property values used in the analysis is provided in ASCE 4. Selected methods shall be approved by the engineer of record.

Neg.= 0

On the first Paragraph, second and third lines: The seismic level for the site (whether located in low or high seismic region) should not be the basis of selecting linear or nonlinear seismic analysis. As we know that the tank system must be designed to maintain continuous operation during and after OBE event, and thus, the linear elastic approach (R=1) has been commonly applied. Thus, suggest the following: "In general, linear seismic analysis is used for OBE case, while non-linear analysis is used for SSE case."

Wu analysis is used in regions with higher seismicity and/or SSE case. When seismic forces are determined using a linear elastic approach, the response modification factor should be taken as R=1. A force reduction factor of R > 1 may be used if it can be demonstrated explicitly shown, by means of dynamic or static (e.g., “push-over”) nonlinear analysis, that the structure meets or exceeds the performance criteria prescribed in this Code. Guidance for selecting modeling methodologies, material properties, and other values used in the analysis is provided in ASCE 4. Selected methods shall be approved by the engineer of record. These are only editorial changes - unanimously agreed by all the TG members present: Brannan, Hjorteset, Ballard, NKO.

document should be used as a guideline for analysis practices that produce reliabable seismic demands for design.

ACI 376 / 376 R Last Update: 3/20/2008 Chapter VI – Standard and Commentary


CHAPTER 6 – ANALYSIS AND DESIGN

Comments for Editorial TG

CODE Commentary

6.1 – Methods of analysis

R6.1 – Methods of analysis

6.1.1 – Required analysis - The containment structure shall be analyzed as an integrated structure that includes the foundation, wall, roof, contained liquid, liner or portion of the liner that is assumed to act compositely with the concrete structure. The effects of discontinuities shall be considered. Constitutive models, assumed values and details used in the analysis shall be approved by the owner/engineer.

R6.1.1 – Required analysis - The analysis of imposed mechanical loads, thermal loads and support configurations that do not vary significantly in the circumferential direction can be analyzed using an axi-symmetric dimensional model. For imposed mechanical loads, thermal loads and support configurations that do vary in the circumferential direction a 3-dimensional or 2-dimensional axi-symmetric harmonic analysis should be performed. Consideration should also be given to the presence of structural discontinuities raising local stresses that are in addition to the global stress fields determined from a 2-D Axi-symetric analysis. In particular in the circumferential direction local to the buttresses and in the vertical direction at the buttress to slab connection. 3-D analysis should be used to determine the effects of post tensioning sequence on the outer tank local to and within the access opening. Emphasis should be placed on the stress state within the access opening due to the absence of self weight in this area and potential failure to attain the performance levels of this standard. Where assumptions are made to simplify the level of analysis; for instance where pile groups are simplified from 3-D orthogonal/radial behavior to axisymmetric behavior; then verification should be carried out to ensure that the analysis assumptions adequately capture and bound the actual behavior. All temperature variations should be based on 95th and 5th percentile temperatures, additionally the corresponding effects of solar radiation should be incorporated within all thermal related analyses including those for normal and spillage load cases. Stress free temperatures should be taken as upper and lower bound within the analysis adequately reflecting the construction period and historical data. In this respect a heat transfer analysis should be undertaken and the film coefficients determined based on the object size and flow conditions. Film coefficients should be correlated to the surface temperatures of the tank. Unless otherwise specified, the vertical tank should be considered as a cylinder in cross flow subjected to a wind speed of 4 m/s. The roof should be considered as a flat plate with due allowance for the effects of the dome shape again in a flow



CODE Commentary of 4 m/s. A minimum wind speed of 4 m/s is a value historically used in the design. For solar radiation and temperature loading a 2-dimensional axisymetric model is sufficient for determination of global loads. The cracking analysis should be based on a Finite Element Method that (1) uses recognized or codified constitutive models for the stress strain behavior of concrete, and (2) incorporates tension stiffening effects. When calculating crack widths the tension stiffening term should not be deducted from the calculation where tension stiffening is explicitly included in the analysis. Additionally the crack widths should be calculated as characteristic and not mean crack widths. Unless otherwise specified, the concrete constitutive mode from European Code EC2, should be used. In this case, the crack widths should be calculated as characteristic and not mean crack widths. The effect of soil stiffness should be included in the analysis as defined in 6.1.2. For load conditions 3.1.15 and 3.1.16, which include severe thermal loading conditions, the structure should be analyzed for the entire transient history up to and including steady state. Both maximum and minimum design ambient temperatures should be used as the initial temperature profiles for the analysis of all loading conditions. For load condition 3.1.15, the temperature for the entire structure, including the roof, should be estimated. The temperature resulting from vapor generation during roll-over and spill events shall be taken into account. The structural model for load conditions 3.1.15 and 3.1.16 should consider the entire temperature time history and be analyzed on the basis of transient inelastic response. Serviceability requirements should be checked both during the transient and steady state temperature profiles. The analysis for these thermal load conditions should take into account the effect of cracking and tension stiffening. Cracking and tension stiffening should be included by appropriate modification of the material stress strain relationship or by the use of finite elements that have the capability of cracking under tension, and crushing under compression as well as the ability to include reinforcing steel. Note for the Editorial TG: Include original references for the model used by the EC2 code. Also include full reference to EC2, including listing it in the reference section.

6.1.2 – Soil and Pile Stiffness - For any analysis of R6.1.2 The range of soil properties to be used in the



CODE Commentary the structure that includes either the static soil / pile stiffness (short-term or long-term settlement) or dynamic soil / pile stiffness, the analysis shall include a practical lower and upper bound range of soil properties. The range of soil stiffness shall be included as part of the Geotechnical Investigation and determined by the Geotechnical Engineer. When established by the Geotechnical Investigation non-linear soil properties and/or non-linear pile stiffness shall be included in the static and dynamic analysis of the structure. The range of values is to be defined by the Geotechnical Engineer based in the soils investigation therefore further guidance in the provisions is not included. Some guidance is included in the commentary

analysis is not intended to be an absolute maximum range but a range that as a result of the subsurface investigation reasonably brackets the properties of the soil strata. The more extensive the subsurface investigation and/or uniformity of the subgrade may reduce the range of values to be used in the analysis. The short-term or long-term settlement should be included in the analysis as deformations consistent with the loadings used to develop the settlement profile. For slab on grade foundations, the range of dynamic soil stiffness need not exceed twice the mean value for the upper bound or one-half the mean value as a lower bound. For pile foundations, to account for the variable soil properties and mechanism for developing resistance, an equivalent range of 100% greater than the unfactored stiffness and a lower bound of 50% of the unfactored stiffness is accepted practice. [Refer: “Recommended LRFD Guidelines for the Seismic design of Highway Bridges, Part I: Specifications”, Prepared under the MCEER Highway Project, Project 94, Task F3-1, November 2001]

6.1.3 6.1.2 - Seismic Analysis

R6.1.3 R6.1.2 - Seismic Analysis

6.1.3.1 6.1.2.1 – General The seismic analyses of the RLG tank foundation system shall be performed for the OBE, SSE and SSEaft events. The effect of tank wall flexibility shall be considered in these analyses. The reduction of responses due to soil - structure interaction (SSI) effects shall be permitted, but limited to a maximum reduction of 50% for SSE analysis and 40% for SSEaft and OBE analyses.

R6.1.3.1 R6.1.2.1 – General – Both linear and non-linear analysis may be used in determining seismic forces. In general, linear analysis is used in the case of low seismic regions and/or OBE case, while non-linear analysis is used in regions with higher seismicity and/or SSE case. When seismic forces are determined using a linear elastic approach, the response modification factor should be taken as R=1. A force reduction factor of R > 1 may be used if it can be demonstrated explicitly shown, by means of dynamic or static (e.g., “push-over”) nonlinear analysis, that the structure meets or exceeds the performance criteria prescribed in this Code. Guidance for selecting modeling methodologies, material properties, and other values used in the analysis is provided in ASCE 4. Selected methods shall be approved by the engineer of record.

6.1.3.2 6.1.2.2 – Seismic Analysis Methods - The response spectra or time history analysis method shall be used for calculating the seismic responses of the tank-fluid-foundation system. Both horizontal and vertical ground motions, defined either as response

R6.1.3.2 R6.1.2.2 – Seismic Analysis Methods - The modal superposition method is used for response spectra analysis. For time history analysis, the modal superposition or direct integration method can be used for calculating the seismic responses (Ref. 8). The time histories should meet



CODE Commentary spectra or time histories, shall be considered in the seismic analysis.

the amplitude, frequency and duration requirements for the site for OBE, SSE and SSEaft events. When the tank is located in high seismic area, and is susceptible to partial uplifting at the base, the seismic analysis may include the nonlinear effect due to base uplifting (Ref. 15).

6.1.3.3 6.1.2.3 – Finite Element Model of Tank-Fluid-Foundation System The finite element model of tank-fluid foundation system shall include the liquid content, inner tank, outer tank, roof, and soil/pile foundation (for SSI effects). When the analysis is performed without the SSI effect, the tank-fluid model shall be permitted to be constructed in accordance with ACI 350.3 (Ref. 4).

R6.1.3.3 R6.1.2.3 – Finite Element Model of Tank-Fluid-Foundation System - The member axial, bending and shear stiffnesses are used to construct the detailed finite element or stick models of the tank-fluid-foundation system. The steel roof and suspended deck, where applicable, should be modeled with the outer tank to account for the dynamic amplification of the vertical accelerations. Detailed procedures for developing the stick model are discussed in Ref. 2 to 7. The impulsive and convective masses with the associated spring constants are lumped at appropriate heights on the inner tank stick model. The hydrodynamic forces due to seismic excitation are the combination of the impulsive and convective forces. For determining the dynamic foundation impedances for the SSI analysis, strain-compatible dynamic soil properties shall be used (Ref. 8 and 19). Service from the Geotechnical Consultant is required.

6.1.3.4 6.1.2.4 – Damping Consideration and Seismic Analysis - The seismic analysis of the tank-fluid-foundation system shall take into account damping expressed as a percentage of critical damping. Types of damping considered in seismic analysis shall include structural damping, convective damping, foundation damping (in conjunction with an SSI analysis) and system (or composite modal) damping:

a) Structural Damping: This damping is related to the type of tank material. Since the impulsive liquid moves with the structure, impulsive damping is a type of structural damping. Structural damping values provided in the following table shall be used unless higher values can be justified through tests or reference.

Tank Type OBE SSE Reinforced Concrete 4% 7% Prestressed Concrete 2% 5% Steel 2% 4%

b) Convective (fluid) damping: This damping is

R6.1.3.4 R6.1.2.4 - – Damping Consideration and Seismic Analysis - The structural damping values are extracted from Reference 22. When the tank foundation can be considered as fixed-base (shear wave velocity ≥ 2500 fps), only the structural damping values are used in the seismic analysis. The foundation radiation damping is a function of the excitation frequency. When the foundation soil medium is relatively uniform (similar to the elastic half-space), the foundation damping can be assumed to be frequency-independent, and can be evaluated based on Ref. 8 and 21. Only the impulsive mode is included in the evaluation of the system damping for a tank-fluid-foundation system. The convective (sloshing) mode that exhibits a very long period of vibration is considered as decoupled mode from the finite element tank-foundation model. For calculating the system damping, various weighting techniques are presented in Ref. 8. The stiffness weighting technique is commonly used. Consideration of the SSI effect will increase the effective vibration period of the tank-fluid-foundation system, and



CODE Commentary associated with sloshing response of the liquid. The damping for convective (sloshing) action shall be 0.5% of critical.

c) Foundation Damping: The radiation and viscous damping for soil/pile foundation must be considered in SSI analysis. The foundation damping for any vibration mode shall not exceed 25% of critical.

d) System Damping: In the SSI model, with different damping values in tank-fluid-foundation system, the system damping (or composite modal damping) needs to be calculated for each vibration mode for determining the dynamic modal responses. The system damping for any vibration modes shall not exceed 15% for OBE and SSEaft, and 20% for SSE

The horizontal and vertical acceleration response spectra defined in Section 3.1.6.1 shall be constructed covering the entire range of anticipated damping ratios and natural periods of vibration, including the sloshing (convective) mode of vibration. The impulsive and convective modal responses shall be combined by the SRSS (square-root-of-sum-of -squares) method. The horizontal and vertical loads shall be combined by the (1-0.3-0.3) rule. The OBE, SSE and SSEaft seismic responses such as accelerations, member forces and moments shall be combined with other applicable static loads, for design of inner and outer concrete tank and foundation. Also, for design of suspended deck, steel roof and other equipment supported at the roof, the maximum seismic acceleration responses at the top of the wall shall be required

generally the overall system damping. Thus, the seismic response will be reduced. A simple and practical approach for calculating the effective vibration period and system damping for SSI consideration is presented in Ref. 11. For a complex dynamic soil-pile-tank foundation interaction problem, the seismic response may be determined based on the finite element seismic analysis method (Ref. 18).

6.2 – Design Basis

R6.2 – Design Basis

6.2.1 – General

R6.2.1 – General

6.2.1.1 Concrete and prestressed concrete containers, associated concrete structures and components of the structures, shall be proportioned to have design strengths at all sections equal to or exceeding the minimum required strengths calculated for the factored loads and forces in such combinations as specified in Chapter 3.

R6.2.1.1 The objective of the container design is to ensure that the container meets all the performance criteria prescribed in Chapter 4, both during service conditions and abnormal load conditions. While the design is primarily based on the Strength Design method, a number of loading conditions and serviceability performance criteria (particularly those associated with abnormal loads) lend themselves to the Allowable Stress Design method.



CODE Commentary

6.2.1.2 – Design of prestressed concrete containers shall be based on strength and on behavior at service conditions at all load stages that will be critical during the life of the structure from the time prestress is first applied.

6.2.1.3 - The design of the concrete and prestressed concrete containment shall be in accordance with the provisions of ACI 350 except as otherwise modified or supplemented in this Standard.

6.2.2 – Required Strength - The required strength to resist the loads specified in 3.1 shall be at least equal to the resultant factored load for the load combinations prescribed in 3.2 combined with the load factors defined in Table 5.2.

6.2.3 – Design strength - The design strength provided by a member or cross section shall be taken as the product of the nominal strength, calculated in accordance with the provisions of this Standard, multiplied by the applicable strength reduction factor specified in Section 5.

6.2.4 – Serviceability requirements The container shall be designed to meet or exceed the serviceability requirements prescribed in Chapter 4.

6.3 – Foundation Design

R6.3 – Foundation Design

6.3.1 – The foundation shall be constructed of concrete with a minimum 28-day cylinder compressive strength of 4000 psi (30 MPa).

6.3.2 – For slab foundations not in contact with RLG and the associated temperatures the slab shall have a minimum thickness of 12 inches (300 mm). The minimum reinforcing, cover and bar spacing shall be in accordance with ACI 350. For slab foundations in contact with RLG shall have a minimum thickness of 12 inches (300 mm). The slab shall have a minimum reinforcing ratio of 0.006 in each direction. The upper mat of reinforcing should be located in the top 3.5 inches (90 mm) and shall have a minimum ratio of reinforcing area to total concrete area of 0.004 in each orthogonal direction. The lower mat of reinforcing shall be located in the bottom 5 inches (130 mm) of the slab and shall have a minimum ratio of reinforcing area to total concrete area of 0.002 in each orthogonal direction.

R6.3.2 –Requirements for the structural foundation slab are different from those for a liquid-tight slab since the secondary bottom provides a leak tight barrier that protects the slab from the effects of the spilled product.



CODE Commentary The maximum bar spacing shall not exceed 12 inches (300 mm) and the minimum bar size shall be #4. No less than 1/3 of the required area of shrinkage and temperature steel shall be distributed at any one surface. 6.3.3 – Structural slabs and pile caps shall be designed and detailed in accordance with ACI 350. Minimum reinforcing requirements of 6.3.2 shall be also be included in the design.

6.3.4 – When seismic loads dictate that anchors are required to resist the inner tank seismic overturning loads the slab or pile cap shall be designed to resist the anchor loads. The OBE and SSE anchor loads shall not include any inelastic behavior of the inner tank, inner tank anchors or other components that reduce the anchor loads.

R6.3.4 The pullout capacity of the anchor, the flexural resistance of the slab and pile cap punching shear must be sufficient to insure that the inner tank can, if required, develop an inelastic response. Since pullout and punching shear are brittle failure in nature no credit for ductility is permitted in the design.

6.3.5 – If the slab or pile cap is thickened at the outside circumference additional reinforcing shall be added to maintain the minimum reinforcing ratio.

6.3.6 – If a monolithic wall to foundation joint is incorporated in the design the effect of wall stiffness and forces shall be included in the analysis of the slab for the predicted differential settlements.

6.3.7 - Reinforcing shall be continuous through construction joints in the slab. All reinforcing shall be fully developed. Development lengths and lap lengths shall be in accordance with ACI 350.

6.4 – Wall Design

R6.4 – Wall Design

6.4.1 - The wall shall be constructed of concrete with a minimum 28-day cylinder compressive strength of 5000 psi (35 MPa).

6.4.2 – Non-prestressed reinforcing shall comply with the requirements of Chapter 2 of this Code.

ACI 376 / 376 R Last Update: 3/19/2008 Chapter IX – Construction Requirements

Ballot Results 3/19/08 and TG Houston Meeting Resolution Page 1 of 1

CHAPTER IX – CONSTRUCTION REQUIREMENTS

Section Approved with Editorial Comments resolved

Latest Text Balloted Vote Committee Members’ COMMENTS Author RESPONSE Notes The ballot proposes that the balloted R9.4.1 be replaced with the following text: R9.4.1 - Prestressing ducts should be inspected for blockage before, during and after concrete placement. Corrective action should be developed in advance of the construction.

Ballot Result 2/18 – 3/19/08


How do you inspect it after grouting? Hoff TG Houston Meeting – 3-18-08: Brannan, Hjorteset, Ballard, NKO The TG agrees with the comment. However, the paragraph refers to casting of concrete outside the duct. To clarify this issue the following editorial change is introduced: R9.4.1 - Prestressing ducts should be inspected for blockage before, during and after concrete placement and prior to grouting. Corrective action should be developed in advance of the construction. This is only an editorial change (unanimously agreed by all the TG members present: Brannan, Hjorteset, Ballard, NKO).

ACI 376 / 376 R Last Update: 3/19/08 Chapter IX – Standard and Commentary

Status after 3/19/08 Ballot an March TG Houston Meeting Page 1 of 12

CHAPTER 9 – Construction

Notes for the Editorial TG

CODE Commentary

9.1 - Mockups R9.1 Mockups - Prior to construction, full-scale mockups should be considered to ensure plant equipment and labor force can attain required quality.

9.2 - Tolerances – Tolerances shall be as per ACI 117 and ACI 301. Additional requirements listed in sections 9.2.1 to 9.2.8 shall also be satisfied. Tolerances are not cumulative and the most restrictive tolerances should apply.

9.2.1 - Tolerances for Cross Sectional Dimensions - The variation in cross sectional dimensions of slabs shall be as follows;

a) ±3/8” where the section thickness is less than or equal to 8 in.

b) ±½” where the section thickness exceeds 8 in. The cross sectional dimensions of walls shall not vary by more than ±¼” from the specified thickness. The cross sectional dimensions of dome roof shall not vary by more than ±½” from the specified thickness. TOLERANCES TO BE CONFIRMED IN ACI 350

9.2 - Tolerances – Tolerances shall be as per ACI 117 and ACI 301. Additional requirements listed in sections 9.2.1 to 9.2.8 shall also be satisfied.

9.2.2 – Variation in Roundness - The maximum permissible deviation from the base slab radius measured to the outside face of the base slab shall be the lesser of

a) 0.10% of the base slab radius b) 1.5 in.

Additionally under no circumstances shall the tolerance be less than ±¾”. The maximum permissible deviation from the specified tank wall radius measured to the inside face of the tank wall at the bottom of the wall shall be the lesser of

a) 0.06% of the tank radius

R9.2.2 - Variation in roundness - Tolerances for wall roundness at the bottom of the tank are normally documented as ±1” at base of the wall for large diameter LNG tanks. In order to cover the full spectrum of potential tank diameters and to introduce sensible upper bound values clauses a) and b) have been introduced. ACI 372/373/350 states “maximum permissible deviation from the specified tank radius should be 0.1% of the radius” which could result is tolerance issues out with normal construction practice. i.e., for a 130ft radius tank the deviation would be 1.5”. (Normally this would be 1”). For alignment the tolerance has been set at .06% of the tank radius which conforms to existing practice (40m radius tank



CODE Commentary b) 1 in.

Under no circumstances shall target deviation be less than ±½”. The maximum permissible deviation from the specified tank wall radius measured to the inside face of the tank wall at the top of the wall shall be the lesser of

a) 0.10% of the tank radius b) 1.5”

Under no circumstances shall the target deviation be less than ±¾”.

Intermediate tolerances between the bottom and top of the wall may be interpolated.

=25mm deviation). Due to the incorporation of 0.06%xR the tolerance will float between lower and upper bound limits for tanks in the radius range of 69.5’ and 138’. Tanks smaller than 69.5’ will have a permitted deviation of ±½”. Similarly tanks larger than 138’ will have a permitted deviation of ±1” Deviations at the top of the tank reflect existing construction practice and facilitate a range of tank diameters.

9.2.3 - Localized Tank Radius - The maximum permissible deviation of the tank wall radius measured along any 10 ft. of circumference shall be 10% of the wall radius. The maximum permissible deviation of the base slab radius measured along any 10 ft. of circumference shall be 20% of the base slab radius.

9.2.4 - Vertical Walls a) Walls shall be plumb within 1/4” per 10 ft of vertical

dimension. b) The variation is verticality measured from the

bottom of the wall to the point of consideration shall not be more than 1”.

9.2.5 - Level Alignment a) The variation in level from specified elevations for

any completed surface excluding the support underneath the inner tank wall shall be limited to ±½”.

b) Along any 10 ft straight line the difference in elevation of any two points shall not exceed ±3/16”.

c) In the instance of the edge of the base slab, and walls along any 10 ft circumference the difference in elevation of any two points shall not exceed ±3/16”.

d) The variation in level from specified elevations for any completed concrete surface supporting the inner tank and annular plates shall not exceed ±¼” in the total circumference.

The top of the concrete layer supporting the inner tank shall be level within ±1/8” in any 30 ft circumference and



CODE Commentary within ±¼” in the total circumference. 9.2.6 - Miscellaneous embedments and openings

a) Location of the centerline of access openings and cross sectional dimensions of access openings shall be within ±½”.

b) Variation in the position of cast in insert plates measured in the plane of concrete surfaces ±½”.

c) Variation in the position of cast in insert plates measured normal to the plane of concrete surfaces ±3/16”.

d) Variation in the position of openings and sleeves shall not exceed ±¼”.

R9.2.6 – Miscellaneous embedments and openings. Positional alignment of insert plates for typical attachments such as ladders and pipe support may be relaxed by the Engineer provided that any positional tolerances are considered on the sizing and selection of the plate size.

9.2.7 - Vertical Vapor Barrier Embedment Tolerances a) The position of vertical liner embedments

measured at the bottom of the wall in the plane of the concrete surface shall be shall ±¼”.

b) Vertical liner embedments shall be plumb within ±½” when measured from the bottom to the top of the wall.

The variation in position of vertical liner embedments measured normal to the concrete surface shall be +0”, -1/8”.

R9.2.7 - Vertical Vapor Barrier Embedment Tolerances - Negative tolerance infers recessed embedments under such conditions that the concrete adjacent to the embedment should be sloped to avoid step changes and facilitate vapour barrier connection.

9.2.8 - Horizontal Thermal Corner Protection Embedments

a) The elevation of horizontal thermal corner embedments shall be within ±¼”.

b) The variation in position of horizontal embedments measured normal to the concrete surface shall be +0”, -1/8”.

R9.2.8 - Horizontal Thermal Corner Protection Embedments - Negative tolerance infers recessed embedments under such conditions that the concrete adjacent to the embedment should be sloped to avoid step changes and facilitate vapour barrier connection.

9.3 - Shotcrete - Unless otherwise indicated here, shotcrete shall meet the requirements of ACI 506.2.

9.3.1 - Proportioning Shotcrete - Shotcrete shall be proportioned in accordance with the following requirements :

a) Wire coat shall consist of one part Portland cement and not more than three parts fine aggregate by weight.

b) Body coat shall consist of one part Portland cement and not more than four parts fine aggregate by weight.

c) Proportioning shall provide a 28-day minimum compressive strength of shotcrete not less than 4,500 psi.

9.3.2 - Shotcrete Construction Procedures - Procedures



CODE Commentary for shotcrete construction of primary and secondary containers comprising circular wire and strand shall be as specified in ACI 506.2 except as modified herein. 9.3.3 - Shotcrete Overcoat

9.3.3.1 - Externally applied circumferential prestressed reinforcement shall be protected against corrosion and other damage by a shotcrete overcoat.

R9.3.3.1 - The shotcrete covercoat generally consists of two or more coats: a wire coat placed on the pre-stressed reinforcement, and a bodycoat placed on the wire coat. If the covercoat is placed in one coat, the mixture should be the same as the wire coat.

9.3.3.2 - Each layer of circumferential prestressed wire or strand shall be covered first with a wirecoat of cement mortar applied by the pneumatic process as soon as practical after prestressing. The shotcrete shall be wet, but not dripping, and provide a minimum cover over the wire of ¼ in. The nozzle shall be held at a small upward angle not exceeding 5 degrees and shall be constantly moving, and always pointing in a radial direction toward the center of the tank. The nozzle shall deliver a steady, uninterrupted flow of shotcrete. The nozzle distance from the prestressed reinforcement shall be such that shotcrete does not build up over or cover the front faces of the wires or strands until the spaces between them are filled.

R9.3.3.2 - Nozzle distance and wetness of mixture are equally critical to satisfactory encasement of pre-stressed reinforcement. If the nozzle is held too far back, the shotcrete will deposit on the face of the wire or strand at the same time that it is building up on the core wall, thereby not filling the space behind them. This condition is readily apparent and should be corrected immediately by adjusting the nozzle distance and, if necessary, the water content.

9.3.3.3 - The wire coat shall be damp-cured by a constant spray or trickling of water down the wall, except that curing shall be permitted to be interrupted during continuous prestressing operations. Curing compounds shall not be used on surfaces that will receive additional shotcrete.

R9.3.3.3 - Curing compounds applied to intermediate layers of shotcrete may interfere with the bonding of subsequent layers and thus their use is prohibited. Maintaining the relative humidity, naturally or artificially, near or above 95% over the shotcrete surface is an acceptable method of curing in accordance with ACI 506R.

9.3.3.4 - Shotcrete material placed incorrectly shall be removed and replaced.

R9.3.3.4 - After the wire coat is in place, visual inspection can immediately determine whether or not proper encasement has been achieved. Where the reinforcement patterns show on the surface as distinct continuous horizontal ridges, the shotcrete has not been driven behind the reinforcement and voids can be expected. If, however, the surface is substantially flat and shows virtually no pattern, a minimum of voids is likely.

9.3.3.5 - A body coat providing a minimum of 1 in. cover over the outside layer of prestressed reinforcement shall be applied over the last layer of wire coat.

• If the body coat is not applied as a part of the wire coat, laitence and loose particles shall be removed from the surface of the wire coat prior to the application of the body coat.

• Thickness control shall be as required by ACI 506.2.

• The completed shotcrete coating shall be cured

R9.3.3.5- Curing should be started immediately after shotcrete placement without damaging the shotcrete.



CODE Commentary for at least 7 days using methods specified by ACI 506.2.

9.3.3.1 - After the bodycoat has cured, the surface shall be checked for “hollow sounding” or “drummy” spots by tapping with a light hammer or similar tool. Such spots indicate a lack of bond between coats and shall be repaired. These areas shall be repaired by removal and replacement with properly bonded shotcrete, or by epoxy injection.

9.3.4 - Thickness control of shotcrete core walls and covercoats

R9.3.4 - Thickness control of shotcrete core walls and covercoats

9.3.4.1 - Positive methods shall be used to establish uniform and correct thickness of shotcrete core.

R9.3.4.1 - Vertical screed wires are the normal method used to establish uniform and correct final thickness of shotcrete and should be spaced not more than 36in apart circumferentially. Wires should be installed under tension, defining the outside surface of the shotcrete from top to bottom. Wires generally are 18- to 20-gauge high-tensile-strength steel wire. Other methods may be used that will provide positive control of the thickness

9.4 - Post Tensioning

R9.4 - Post Tensioning

9.4.1 - Ducts for grouted tendons shall be mortar-tight and non reactive with concrete, tendons, or filler material.

R9.4.1 – Prestressing ducts should be inspected for blockage before, during and after concrete placement and prior to grouting. Corrective action should be developed in advance of the construction.

9.4.2 - Ducts for grouted single wire, strand, or bar tendons shall have an inside diameter at least +¼ in. larger than tendon diameter.

9.4.3 - Ducts for grouted multiple wire, strand, or bar tendons shall have an inside cross-sectional area at least two times area of tendons.

9.4.4 - Ducts shall be maintained free of water if members to be grouted are exposed to temperatures below freezing prior to grouting.

9.4.5 - Vent holes shall be provided adjacent to the anchorages and at high points of the ducts to assist with air removal.

R9.4.5 - Vent holes shall be provided at anchorage locations at high points and at positions recommended by the Post tensioning supplier to ensure proper grouting quality.

9.4.6 – Vertical Tendons - Vertical tendon shall be filled



CODE Commentary from the bottom. 9.4.6 – Grout for Bonded Prestressing Tendons - Grout trials shall be used to ensure the following;

• The grout has not shrunk away from the duct wall or strands whereby creating voids within the system

• The grout has the correct flow characteristics to reach each of the vent points (where provided) and to completely fill the duct.

• The theoretical grout consumption

R9.4.6 – Grout for Bonded Prestressing Tendons - Full scale grout tests are performed on the horizontal tendons to specifically demonstrate the selection of tendon and distribution of vent tubes and selection of pumping equipment. After the grout trials the tendon is cut transversely and inspected prior to commencement of grouting operations proper. Grout provides bond between the post-tensioning tendons and the concrete and by which corrosion protection of the tendons is assured. Proper grout and grouting procedures, therefore, play an important part in post-tensioned construction. Reference PT manual. Past success with grout for bonded prestressing tendons has been with Portland cement as the cementing material. A blanket endorsement of all cementitious materials (defined in Chapter II) for use with this grout is deemed inappropriate because of a lack of experience or tests with cementitious materials other than Portland cement and a concern that some cementitious materials might introduce chemicals listed as harmful to tendons in R18.16.2. Thus, “Portland cement” in 18.16.1 and “water-cement ratio” in 18.16.3.3 are retained in this edition of the code.

9.4.7 - Grout shall consist of Portland cement and water; or Portland cement, sand, and water; or a 100 % solids, two-component epoxy resin system.

R9.4.7 - Epoxy grout has been used in limited applications. Caution is recommended in its selection and use. Properties of the material should be reviewed including differences in the coefficient of thermal expansion and heat generation.

9.4.8 - Grout material shall conform to the following requirements:

a) Cement for grouting operations shall be Type I or Type II in accordance with ASTM C150.

b) The minimum compressive strength of grout shall be 6000psi at 28 days tested in accordance with ASTM C109.

c) Water cement ratio shall not exceed 0.45 by weight of cement.

d) Sand, if used, shall conform to “Standard Specification for Aggregate for Masonry Mortar” (ASTM C 144) except that gradation shall be permitted to be modified as necessary to obtain satisfactory workability.

e) Admixtures conforming to (ASTM C494) and known to have no injurious effects on grout, steel, or concrete shall be permitted. Calcium chloride shall not be used.

f) Bleeding of the grout shall not exceed 2% of the volume for the first 3 hours after mixing, nor 4% total at any time. All separated water shall be

R9.4.8 - The limitations on admixtures in ASTM C494 apply to grout. Substances known to be harmful to prestressing tendons, grout, or concrete are chlorides, fluorides, sulfites, and nitrates. Aluminum powder or other expansive admixtures, when approved, should produce an unconfined expansion of 5 to 10 percent. Neat cement grout is used in almost all building construction. Only with large ducts having large void areas should the advantages of using finely graded sand in the grout be considered. Admixtures are generally used to increase workability, reduce bleeding and shrinkage, or provide expansion. This is especially desirable for grouting of vertical tendons.



CODE Commentary reabsorbed within 24 hours.

g) Epoxy grout shall be moisture insensitive with a minimum compressive strength of 125 percent of the design concrete compressive strength.

9.4.9 – Grout Proportions

R9.4.9 – Grout Proportions

9.4.9.1 - Proportions of materials for grout shall be based on either of the following:

a) Results of tests on fresh and hardened grout prior to beginning grouting operations, or

b) Prior documented experience with similar materials and equipment and under comparable field conditions.

R9.4.9.1 - Grout proportioned in accordance with these provisions will generally lead to 7-day compressive strength on standard 2-in. cubes in excess of 2500 psi and 28-day strengths of about 4000 psi. The handling and placing properties of grout are usually given more consideration than strength when designing grout mixtures. Hoptay to resolve Comment (see Puerto Rico notes): R9.4.9.1 indicates that grout proportioned in accordance with these provisions will produce 28 day compressive strengths of about 4000 psi. This is inconsistent with 9.4.8 which requires a minimum strength of 6000 psi.

9.4.9.2 - Cement used in the work shall correspond to that on which selection of grout proportions was based.

9.4.9.3 - Water content shall be minimum necessary for proper pumping of grout; however, water cement ratio shall not exceed 0.45 by weight.

9.4.9.4 - Water shall not be added to increase grout flowability that has been decreased by delayed use of grout.

9.4.9.5 - Epoxy grout shall have demonstrated by tests or experience to exhibit acceptable pumpability and low exothermic

9.4.10 – Grout Mixing and Pumping

R9.4.10 – Grout Mixing and Pumping - In an ambient temperature of 35 F, grout with an initial minimum temperature of 60 F may require as much as 5 days to reach strength of 800 psi. A minimum grout temperature of 60 F is suggested because it is consistent with the recommended minimum temperature for concrete placed at an ambient temperature of 35 F. Quickset grouts, when approved, may require shorter periods of protection and the recommendations of the suppliers should be followed. Test cubes should be cured under temperature and moisture conditions as close as possible to those of the grout in the member. Grout temperatures in excess of 90 F will lead to difficulties in pumping

9.4.10.1 – Grout shall be mixed in equipment capable of continuous mechanical mixing and agitation that will produce uniform distribution of materials, passed through screens, and pumped in a manner that will completely fill tendon ducts.



CODE Commentary 9.4.10.2 – Temperature of members at time of grouting shall be above 35 F and shall be maintained above 35 F until field-cured 2-in. cubes of grout reach a minimum compressive strength of 800 psi.

9.4.10.3 – Grout temperatures shall not be above 90 F during mixing and pumping.

9.4.11 – Protection of Prestressing Tendons - Burning or welding operations in vicinity of prestressing tendons shall be carefully performed, so that tendons are not subject to excessive temperatures, welding sparks, or ground currents.

9.4.12 – Application and Measurement of Prestressing Force

R9.4.12 – Application and Measurement of Prestressing Force

9.4.12.1 – Prestressing force shall be determined by both of the following methods:

a) Measurement of tendon elongation. Required elongation shall be determined from average load-elongation curves for the pre-stressing tendons used.

b) Observation of jacking force on a calibrated gage or load cell or by use of a calibrated dynamometer.

Cause of difference in force determination between a) and b) that exceeds 5 percent for pretensioned elements or 7 percent for post tensioned construction shall be ascertained and corrected.

R9.4.12.1 - Elongation measurements for prestressed elements should be in accordance with the procedures outlined in the “Manual for Quality Control for Plants and Production of Precast and Prestressed Concrete Products,” published by the Precast/Prestressed Concrete Institute.18.24 ACI 318-89, 18.18.1, was revised to permit 7 percent tolerance in tendon force determined by gage pressure and elongation measurements for post-tensioned construction. Elongation measurements for post-tensioned construction are affected by several factors that are less significant, or that do not exist, for pretensioned elements. The friction along post-tensioning tendons may be affected to varying degrees by placing tolerances and small irregularities in profile due to concrete placement. The friction coefficients between the tendons and the duct are also subject to variation. The 5 percent tolerance that has appeared in the code since ACI 318-63 was proposed by ACI-ASCE Committee 423 in 1958,18.3 and primarily reflected experience with production of pretensioned concrete elements. Since the tendons for pretensioned elements are usually stressed in air with minimal friction effects, the 5 percent tolerance for such elements was retained.

9.4.12.2 – Where transfer of force from bulkheads of pretensioning bed to concrete is accomplished by flame cutting prestressing tendons, cutting points and cutting sequence shall be predetermined to avoid undesired temporary stresses.

9.4.12.3 – Long lengths of exposed pretensioned strand shall be cut near the member to minimize shock to concrete.

9.4.12.4 – Total loss of prestress due to unreplaced broken tendons shall not exceed 2 percent of total prestress.

R9.4.12.4 – This provision applies to all prestressed concrete members. For cast-in-place post-tensioned slab systems, a “member” should be that portion considered as an element in



CODE Commentary the design, such as the joist and effective slab width in one-

way joist systems, or the column strip or middle strip in two-way flat plate systems.

9.4.13 – Prestressing Sequence

R9.4.13 – Prestressing Sequence

9.4.13.1 – The stressing sequence shall developed so as to minimize the amount of bending and shear stresses within the tank wall.

R9.4.13.1 – The stressing sequence is of significant importance for vertical cylindrical tanks due to the presence of openings and potential for large shear and bending. Therefore it is recommended that the vertical tendons are stressed prior to the hoop tendons to provide a level of pre compression to mitigate any vertical bending and shear. Secondly the hoop tendons should be stressed in such a manner that a prestressed distribution close to the design distribution is achieved. This is normally achieved through stressing the “even” tendons followed by the “odd” tendons (or vice versa). Additionally the design should specifically address the stressing sequence in the vicinity of the access opening.

9.4.13.2 – Tendon sequencing shall be considered in the design of the wall and foundation.

R.9.4.13.2 – Typically the vertical tendons with the exception of those passing through the access opening should be stressed prior to the horizontals tendons in order to attain maximum shear capacity and bending resistance.

9.4.14 – Anchorages - Horizontal anchorages shall be qualified for the product service temperatures.

R9.4.14 – Test results for specific anchorages should be obtained from post tensioning suppliers. Where records of tests cannot be furnished then scale testing of the proposed anchorages should be undertaken. Refer to FIP SR88/2 for additional information.

9.5 – Winding of Prestressed Reinforcement-Wire or Strand

R9.5 – Winding of Prestressed Reinforcement-Wire or Strand

9.5.1 – Qualifications - The stressing system used shall be capable of consistently producing the specified stress at every point around the wall within a tolerance of + 7 % of the specified initial stress in each wire or strand, as specified in ACI 506.2.

R9.5.1 – Qualifications - Winding should be under the direction of a supervisor having technical knowledge of prestressing principles and experience with the winding system being used.

9.5.2 – Anchorage of Wire or Strand - Each coil of prestressed wire or strand shall be anchored to adjacent wire or strand, or to the wall surface, at sufficiently close intervals to minimize the loss of prestress in case of a break during wrapping. Anchoring clamps shall be removed wherever cover over the clamp in the completed structure would be less than 1 in.

9.5.3 – Splicing of Wire or Strand - Ends of individual coils shall be joined by mechanical splicing devices



CODE Commentary qualified for service temperatures and shall develop the specified tensile strength of the reinforcement. 9.5.4 – Concrete or Shotcrete Strength - Concrete or shotcrete strength at time of stressing shall be at least 1.8 times the maximum initial stress due to prestressing in any wall section.

9.5.5 – Stress Measurement and Wire or Strand Winding Records

9.5.5.1 – A calibrated stress-recording device that can be readily recalibrated shall be used to determine stress levels in prestressed reinforcement throughout the wrapping process. At least one stress reading for every coil of wire or strand, or for each 1000 lb thereof, or for every foot of wall per layer, shall be taken after the prestressed reinforcement has been applied on the wall.

R9.5.5.1 – Readings of the force in the prestressed reinforcement inplace on the wall should be made when the wire or strand has reached ambient temperature. All such readings should be made on straight lengths of prestressed reinforcement. A written record of stress readings, including location and layer, should be maintained. This submission should be reviewed prior to acceptance of the work. Continuous electronic recordings taken on the wire or strand in a straight line between the stressing head and the wall may be used in place of the above when the system allows no loss of tension between the reading and final placement on the wall.

9.5.5.2 – The total initial prestress force measured on the wall per vertical foot of height shall be not less than the specified initial force in the locations indicated on the deign force diagram and not more than 5 percent greater than the specified force.

9.6 – Forming

9.6.1 – Slipforming – Slipforming shall meet the requirements of ACI 347.

R9.6.1 – Slipforming is a desirable form of construction where a metallic vapor barrier is not used achieve liquid tightness due to the minimization of horizontal construction joints.

9.6.2* – Planning of Slipforming Operations – Since the slipform is a continuous procedure, planning shall be made for the supply of concrete, reinforcement and embedments so that the slipform can continue 24 hours per day without interference.

R9.6.2* – Planning of Slipforming Operations – Slipforming operations involves a large number of people on a limited amount of space, working at different levels simultaneously. This requires significant attention to detail at the planning stage. Therefore, mock-ups and trails of heavy reinforced



CODE Commentary Back up systems shall be incorporated including craneage, equipment for the slipform including additional pumps, hoses, jacks.

structural elements should be considered prior to construction. Prior to pouring concrete in the forms detailed checklist for start-up slipform operations should be completed. Planning should also include workmen requirements, working tasks and responsibilities for the workmen and guiding of the slipform. Consideration should be made for the checking that enough equipment, spare parts and consumable goods are available. Along with this planning, back up solutions should be worked for all possible errors or faults that may occur. Prior to pouring concrete in the forms detailed checklist for start-up slipform operations shall be completed. For vertical slipforming, if the concrete slip is excessively delayed so that the hardened front of the concrete is approaching the top surface of the concrete, consideration should be given to using chemical retarders on the surface to delay the hardening of the surface. The unhardened binder on the surface can then be washed away, leaving a rough surface for the bonding when the slip is resumed

9.6.2 – Casting and Consolidation – Casting and consolidation shall be in accordance with ACI 309R.

R9.6.2 – Vibrating too deeply will interfere with the setting in the layers below that might cause loose pieces of concrete to occur in the exposed surface underneath the form.

9.7 – Construction Joints

R9.7 – Construction Joints

R9.7 – Construction Joints - Construction joints are defined as concrete surfaces upon or against which concrete is to be placed and to which new concrete is to adhere, that have become so rigid that the new concrete cannot be incorporated integrally by vibration with that previously placed are defined as construction joints. For precast construction the vertical joints should be pumped from bottom to top.

9.7.1 – Location and type of construction joints shall be defined during the design stage.

R9.7.1 –For the integrity of the structure, it is important that all construction joints be carefully defined in construction documents and constructed as required. Any deviations should be approved by the design engineer. Type and details of unplanned joints should also be considered in the design stage.



CODE Commentary

9.7.2 – Surface of concrete construction joints shall be cleaned and laitance removed.

R9.7.2 – The surfaces of concrete at all construction joints should be prepared as called for in ACI 301. Where reliance is placed on concrete alone for leak tightness then the surface of construction joints should be roughened to a minimum amplitude 1/4 in. The final roughened surface shall be clean and all laitance and loose material removed prior to inspection. Prior to placement of new concrete, the final cleaning should be carried out using high-pressure water.

9.7.3 – Immediately before new concrete is placed, all construction joints shall be wetted to be in a saturated surface dry condition.

R9.7.3 – Cleaning with water prior to pouring of concrete avoids the drawing of moisture from the concrete mix and weakening/honeycombing of the joint. Freestanding water should be removed prior to commencement of concreting operations. The requirements of the 1977 ACI 318 code for the use of neat cement on vertical joints have been removed, since it is rarely practical and can be detrimental where deep forms and steel congestion prevent proper access. Often wet blasting and other procedures are more appropriate. Since the code sets only minimum standards the engineer may have to specify special procedures if conditions warrant. The degree to which mortar batches are needed at the start of concrete placement depend on concrete proportions, congestion of steel, vibrator access, and other factors.

9.7.4 – Construction joints shall be so made and located as not to impair the strength of the structure. Provision shall be made for transfer of shear and other forces through construction joints. See ACI 350 paragraph 11.7.9.

R9.7.4 –Construction joints should be located where they will cause the least weakness in the structure. When shear due to gravity load is not significant, as is usually the case in the middle of the span of flexural members, a simple vertical joint may be adequate. Lateral force design may require special design treatment of construction joints. Shear keys, intermittent shear keys, diagonal dowels, or the shear transfer method of ACI 350 paragraph 11.7 may be used whenever a force transfer is required.

9.8 – Design of Formwork – Formwork shall be designed and made as per ACI 350 Chapter VI.

9.10 – Concrete Emdedments – Concrete embedments should be considered and detailed during the design stage. Requirements of ACI 350 shall be satisfied.

ACI 376 / 376 R Last Update: 3/19/08 CHAPTER 10

Status after Ballot 2/18 to 3/19/2008 and TG Houston Meeting Page 1 of 2

CHAPTER 10 – COMMISSIONING AND DECOMISSIONING CRITERIA

Approved Sections and Approved Sections with resolved Editorial Comments Section Approved with Comments to be resolved Negative Vote

CODE Vote Comments Author RESPONSE Notes

This doesn't tell you what should be performed, only when it should be performed. Seems like something was left out of this clause.

Hoff

The paragraph states that " unless otherwise specified the following shall be performed" but the following are not actions but points of time during construction. The paragraph needs to be reworded.

Hoptay

The ballot proposes that the paragraph 10.2.x be replaced with the following text: 10.2.x - Anchorage – Where anchorage is provided that requires tightening of individual anchors, tightening shall be in accordance with procedures defined by the designer. Unless otherwise specified, the following shall be performed: (a) prior to the pneumatic testing of the secondary container, and (b) during the hydrotest, with the primary tank filled at the maximum

water level. Anchorages shall be visually inspected prior to and after testing.



Is "tightening" the best word? In post-tensioning we typically say "tensioning" or "stressing".

Hanskat

TG Houston Meeting – 3-18-08: Brannan, Hjorteset, Ballard, NKO • Hoff’s and Hoptay’s negative votes found persuasive. The

negative is addressed by inserting “anchor tightening the following”, as shown below.

• Hanskat’s editorial comments: TG believes that word “tightening” better reflects the intent which is to at least tighten the connection vs. prestressing or tensioning it. Ultimately, it is the Designers task to choose tightening or prestressing.

Find out whether Hoff and Hoptay agree with the change and whether this addresses their negative. The change is only editorial. Unanimously agreed by all the TG members present: Brannan, Hjorteset, Ballard, NKO. 10.2.x - Anchorage – Where anchorage is provided that requires tightening of individual anchors, tightening shall be in accordance with procedures defined by the designer. Unless otherwise specified, anchor tightening the following shall be performed: (ab) during the hydrotest, with the primary tank filled at the

maximum water level, and (ba) prior to the pneumatic testing of the secondary container,

and. Anchorages shall be visually inspected prior to and after testing.

Ballard: I think (b) and (a) should be reversed since the hydro test should be performed first to insure that the inner tank anchorages are tight before the pressure test. Someone might read this paragraph as a do (a) then do (b) direction.

The ballot proposes that the following text be added to paragraph R10.2.2 R10.2.2 – (see current R10.2.2)

• ASTM G16 and ASTM G46 may could be used as guidelines for determining which pitting and corrosion testing methodology is appropriate for the examination of pitting and corrosion of the surfaces in question after the hydrotest. The procedure to be used, areas to be tested and the acceptable corrosion and pitting limits should be agreed upon by the Engineer, Owner and Contractor before the hydrotest is performed, subject to the criteria of 10.2.1.


Approved = 23 App. w. Com.=1 Abst.= 4 Neg.= 0

From the titles of each of the ASTM standards listed it is not obvious that guidance is provided in either standard as to which testing methodology is appropriate as indicated in R10.2.2.

Hoptay TG Houston Meeting – 3-18-08: Brannan, Hjorteset, Ballard, NKO ASTM G 46 lists a series of testing methodologies including: • Visual inspection, including metallographic examination • Non-destructive inspections including radiographic, electromagnetic, sonics, and

penetrants, • Mass loss, including pit depth measurement (including metallographic, machining,

micrometer or depth gage, microscopical), • etc. Text was adjusted as shown below. Furthermore, the text is moved to R10.2.2* This is only an editorial change. Unanimously agreed by all the TG members present: Brannan, Hjorteset, Ballard, NKO. R10.2.2 – (see current R10.2.2) ASTM G16 and ASTM G46 may could be used as guidelines for determining which pitting and

ACI 376 / 376 R Last Update: 3/19/08 CHAPTER 10

Status after Ballot 2/18 to 3/19/2008 and TG Houston Meeting Page 2 of 2

CODE Vote Comments Author RESPONSE Notes corrosion testing methodology is appropriate for the examination of pitting and corrosion of the surfaces in question after the hydrotest. ASTM G 16 may be used for applying statistical analysis to corrosion data. The procedure to be used, areas to be tested and the acceptable corrosion and pitting limits should be agreed upon by the Engineer, Owner and Contractor before the hydrotest is performed, subject to the criteria of 10.2.1.

Revise last sentence to read: " Alternatively, in-situ component testing of relief and vacuum breaker valves by applying test gas pressure or vacuum to the control valve pressure/vacuum sensing line is permitted. The set point of controls shall be calibrated with a dead weight tester." Add the following sentence: "Monitor the pressure/vacuum in the tank at all times during testing using instruments with alarm settings to guard against pressure/vacuum conditions outside design limits."

Brannan

Editorial suggestion, change the bullet points to read: - Increasing pressure to ... check pressure relief - Creating a vacuum to ... check vacuum relief

Pawski

The ballot proposes that Hatfield’s negative vote is found convincing and that it be introduced as shown below: 10.3.x – Pressure and Vacuum Relief Testing – Proper functioning of all pressure and vacuum relief valves and devices shall be confirmed by:

• Check pressure relief by increasing pressure in the vapor space. • Check vacuum relief by creating a vacuum with a vacuum pump,

or alternatively, by partially withdrawing water from the tank. Alternatively, in-situ component testing of relief & vacuum breaker valves with test gas applied to the pressure sensing line and set point of controls calibrated with a dead weight tester.



I agree that during normal plant operation that the pressure relief valves are tested as described in the alternate. However, prior to the tank being placed into operation the relief valve sensing system needs to be verified as working properly to insure, for example, that the pressure drop in the sensing line does not cause the valves to reseat prematurely. This is not a new requirement for the testing tank relief valves prior to the tank being placed into service and is a requirement of API 620 for single containment tanks. API 620 Q.9.2.5 states the following," Pressure relief and vacuum relief valves shall be checked by applying the design gas pressure to the outer tank, followed by evacuation of the outer space to the vacuum setting of the relief valve."

Hoptay

TG Houston Meeting – 3-18-08: Brannan, Hjorteset, Ballard, NKO Changes are more than editorial and the paragraph should be reballoted. Proposed changes unanimously agreed to by all the TG members present: Brannan, Hjorteset, Ballard, NKO. 10.3.x – Pressure and Vacuum Relief Testing – Proper functioning of all pressure and vacuum relief valves and devices shall be confirmed by:

• Increasing pressure in the vapor space to check pressure relief by increasing pressure in the vapor space.

• Creating a vacuum to check vacuum relief by creating a vacuum with a vacuum pump, or alternatively, by partially withdrawing water from the tank.

Alternatively, In-situ component testing of relief and vacuum breaker valves by applying with test gas applied to the pressure or vacuum to the control valve pressure/vacuum sensing line is permitted. The and set point of controls shall be calibrated with a dead weight tester. Proper functioning of the relief valves and associated sensing systems shall be verified before the tank is placed into operation. The pressure/vacuum in the tank shall be monitored at all times during testing using instruments with alarm settings to guard against pressure/vacuum conditions outside design limits. Insert in the Commentary R10.3.x - …. Verifying proper functioning of the relief valve system includes, but is not limited to ensuring that the pressure drop in the sensing line does not cause premature reseating of the valves.

ACI 376 / 376 R Last Update: 3/19/2008 Chapter X – Standard and Commentary

Status after 2/18 to 3/19/2008 Ballot and TG Houston Meeting Page 1 of 16

CHAPTER X – COMMISIONING

Comments for the Editorial TG To be balloted

CODE Commentary

10.1 – General Commissioning and decommissioning of primary and secondary containers shall comply with the provisions of this chapter. These provisions shall apply to all the components of the containment systems that are affected by the commissioning operations.

R10.1 – General The term “Commissioning” is used in this section to denote the tests (hydrostatic and pneumatic) that must be conducted prior to placing the tank into service; plus the start-up procedures, such as purging into service and cool-down. “Decommissioning” denotes the purging of the tank out of service, and the subsequent warm-up.

• Concrete primary containers: When the concrete component is part of the primary containment, the testing criteria are nearly similar to those for all-metal primary tanks but with certain modifications that reflect the special properties of concrete.

• Concrete secondary containers only: With few exceptions, when the concrete component is part of the secondary containment only, and 9% Nickel, stainless steel or aluminum plate provide the primary inner liquid containment, the criteria for testing, purging and cool-down and decommissioning procedures.

The provisions of this chapter are intended to highlight commissioning of all cases.

10.2 – Testing

R10.2 – Testing

10.2.1 - General Unless otherwise specified in the contract documents, testing of concrete containers shall include:

• Hydrostatic testing in accordance with Sec. 10.2.2 of primary containers.

• Liquid tightness testing in accordance with Sec. 10.2.4 of primary containers.

• Pneumatic testing in accordance with Sec. 10.2.3 of primary and secondary containers forming a part of the vapor containment.

A written procedure to be followed shall be prepared in advance of testing , to include:

b. Responsibility and duties of operating and supervisory personnel.

c. Safety procedures. d. Test limits. e. Instrumentation installed and monitoring to be

performed. f. Checklists and forms.

Hydrostatic, pneumatic, and liquid tightness testing shall be

R10.2.1 – General Hydrostatic and pneumatic testing is performed to demonstrate structural integrity of the container and its anchorage, bottom insulation, and foundation components prior to filling with RLG. These tests also are part of testing performed to demonstrate liquid tightness. The provisions of this section apply for the hydrotest of the inner or primary containment for RLG tanks. Hydrotesting of secondary containment is not customary, but the provisions of Section u are also applicable for testing the secondary containment, when it is required.



CODE Commentary completed prior to application of insulating materials over surfaces requiring visual inspection during testing. Testing shall not be performed before:

1. Concrete materials have reached specified strength and age.

2. Prestressing installation including grouting and concrete protection is completed.

3. Concrete and reinforcement tests are completed. 4. Inspection and testing of welded joints of metal

liners, penetrations and piping, is completed. 5. Metal surfaces are coated or protected against

corrosion from test water. Testing shall be performed in a manner that does not result in permanent structural or containment deterioration of the tank or metallic components that come into contact with the hydrotest fluid. 10.2.2 – Hydrostatic Testing - The test consists of filling the tank with water to the specified test height and applying an overload air pressure of 1.25 times the pressure for which the vapor space is designed. The water test height is the smaller of:

• Design liquid level of the stored RLG. • Height corresponding to the allowable temporary

test load that may be exerted on foundation components and bottom insulation.

• Height corresponding to the structural limits of any part of the concrete containment

The tank shall be vented to the atmosphere during filling and emptying with water. The period for the complete hydrotest, in general, shall not exceed 30 days, but should be completed as soon as possible. This period is defined as fill, test, empty, wash, and clean and dry time.

R10.2.2 – Hydrostatic Testing

The provisions of this section apply when the inner (primary) container is concrete or metal. Hydrotesting of prestressed secondary containment is not customary. The provisions of Section 10.2 are also applicable to testing of the secondary container, when it is required. For tanks where the inner (primary) container is concrete, the hydrostatic test should be completed before the installation of the wall insulation. Note for the Editorial TG: Concrete secondary containers only: With few exceptions, when the concrete component is part of the secondary containment only, and metal 9% Nickel, stainless steel or aluminum plate provide the primary inner liquid containment is used, the criteria for testing, purging, and cool-down and decommissioning procedures are similar to those for all-metal

10.2.3

10.2.4 – Liquid Tightness - The entire height of a primary container to be exposed to refrigerated liquid including the overfill allowance shall be tested for liquid tightness. Testing for liquid tightness of primary concrete containers without liners shall include one or more of the following:

• Below the hydrostatic test water level visual inspection may be used to check for leaks and wet spots.

• Local pressure/vacuum testing may be used above and below hydrostatic test water levels. Unless otherwise specified in the contract documents local pressure/vacuum testing is required only at construction joints, penetrations, and embedments.

R10.2.4 R10.2.1 – The liquid tightness evaluation can be performed by:

a. filling the primary container and the overfill allowance completely with water, or

b. filling the primary container to an elevation that results in the hydrostatic head equivalence of hydrotest water as compared to the liquefied gas product to be stored in the tank (partial hydrotesting). In some cases, depending on the project requirements, the test hydrostatic head is arrived at by multiplying the equivalent hydrostatic head by a factor ≥ 1.0. In all cases of partial hydotesting, the entire inside surface of the primary container above the hydrostatic test fill line



CODE Commentary • Specified non-destructive testing methods.

Unless otherwise specified in the contract documents testing for liquid tightness of secondary concrete containers is not required.

should be tested with other methods to confirm tightness. These include: i) Local pressure/vacuum testing, such as

external/internal pressure-box testing and internal vacuum box testing as shown in Figures R10.1 to R10.3, or

ii) Non-destructive testing (NDT). The relationship between hydrotesting and NDT techniques should be defined by performing NDT over parts of the tank that are also hydrotested. The supplemental techniques should: 1. Be capable of identifying cracks that are perpendicular to

the wall surface, such as through-thickness cracks, 2. Be capable of identifying other material discontinuities

which adversely affect tank performance, such as presence of voids and/or honeycombing, and

3. Have the ability to cover large areas in a cost-effective and time-efficient manner.

Numerous NDT techniques can be used for evaluating cracking/microcracking and identifying material discontinuities in concrete tanks walls. For instance, impact-echo and ultrasonic pulse-velocity are commonly used for crack location. Ground penetrating radar, impulse response, and ultrasonic pulse velocity are commonly used for detecting honeycombing and voids. Impulse response, impact-echo and ultrasonic pulse velocity are well suited for identifying delamination and debonding. Some of the NDT that seem particularly well suited for identifying through-thickness cracks include Crosshole Sonic Logging (CSL) technique [Olson 1991], and Spectral Analysis of Surface Wave (SASW) technique. It should be noted that defects in the concrete that result in leakage of test water, may self-heal within 3 to 5 days or slightly longer due to the process of autogeneous healing of concrete provided the defect is not too large [Edvardsen 1999]. Defects that do not self-heal will need to be repaired using an appropriate procedure. The test for liquid tightness by hydrotest also determines structural integrity of both the bottom of the wall and the critical wall to base joint. The foundation design is governed by the hydrotest. However, the hoop prestressing requirements are usually not governed by the hydrotest.

10.2.x - Anchorage – Where anchorage is provided that requires tightening of individual anchors, tightening shall be in accordance with procedures defined by the designer. Unless otherwise specified, anchor tightening shall be performed: (a) during the hydrotest, with the primary tank filled at the

maximum water level, and (b) prior to the pneumatic testing of the secondary

container. Anchorages shall be visually inspected prior to and after



CODE Commentary testing.



CODE Commentary 10.2.2* – Quality of Test Water - The test water shall be clean and may include suitable corrosion inhibitors. Use of clean seawater for hydro testing of primary lined or unlined concrete or metal containers is permitted, but at a minimum the following criteria shall be met whether using potable, brackish or seawater for the hydrotest:

• Seawater shall be filtered to remove solids and prohibit introduction of significant quantities of marine life and debris into the tank.

• No hydrogen sulfide is allowed in water. • Water pH shall be between 6 and 8.3. • Water temperature shall be below 120°F (49°C). • For austenitic stainless steel tanks, the chloride

content of the water shall be below 50 parts per million.

• For aluminum tanks, the mercury content of the water shall be less than 0.005 parts per million, and the copper content shall be less than 0.02 parts per million.

• All 9% nickel, stainless steel or aluminum surfaces that will come in contact with seawater shall be adequately protected against corrosion. All weld seams and associated heat-affected zones (HAZ’s) shall be cleaned/prepared and coated with an approved primer after completion of all required NDT inspections. Previously-primed abraded areas shall also be repaired and re-primed.

• All weld seams and associated heat-affected zones (HAZ’s) shall be cleaned/prepared and coated with an approved primer after completion of all required NDT inspections. Previously-primed abraded areas shall also be repaired and re-primed.

• All internal pump columns, stilling wells, standpipes, internal piping, fittings, attachments, guides, etc. shall be 9% nickel or an approved high nickel alloy, except that in case when 9% nickel or high nickel alloys are not available, stainless steel shall be permitted to be used subject to the following limitation: The stainless steel components shall be completely coated on all exposed surfaces with an approved coating, and all inside surfaces shall be sealed during the entire hydrotest cycle. Alternatively, stainless steel components shall be permitted to be installed after the hydrotest.

• The 9% nickel, stainless, or aluminum metal primer shall have proven adhesive performance characteristics suitable for cyclic exposures to cryogenic conditions. Any primer that cannot be demonstrated to have the required adhesion performance shall be stripped after the hydrotest.

• Seawater sampling, corrosion, and pitting tests shall be conducted using the actual seawater from the site prior to hydrotest.

• For metal tanks, or metal components of concrete tanks, the entire surface of the inner tank or component, and all internals exposed to seawater shall be high-pressure spray washed with potable

R10.2.2* – Quality of Test Water - The USEPA Drinking Water Standard has the following limits:

• Chloride limit is less than 250 ppm or mg/l • Copper limit is less than 1 ppm or mg/l • Mercury limit is less than 0.002 ppm or mg/l

Corrosion inhibitors may impact disposal options of test water and should consider local environmental regulations for discharge. The use of seawater as the liquid for the hydrotest in RLG tanks poses a unique set of challenges. Brackish or seawater contains substances that can cause corrosion during the hydrotest if proper precautions are not taken. Furthermore:

• It should be noted that 50ppm is very difficult to achieve.

• The following techniques may be used to provide adequate corrosion protection: prime coating with an approved primer, impressed current cathodic protection, etc. The primer used for coating shall have proven adhesive performance characteristics suitable for cyclic exposures to cryogenic conditions or must be stripped off after the hydrotest. If hydro test water is left in the tank for less than three weeks, 9% Nickel surfaces may be left bare, provided they are thoroughly washed and dried after the hydro test.

In some instances the hydro test water will be maintained in the tank for an extended period to consolidate the soil. It may also be left in the tank to wait for the hydro test of another tank. In these instances, the engineer shall make special provisions.

ASTM G46 may be used as guidelines for determining which pitting corrosion testing methodology is appropriate for the examination of pitting and corrosion of the surfaces in question after the hydrotest. ASTM G 16 may be used for applying statistical analysis to corrosion data. The procedure to be used, areas to be tested and the acceptable corrosion and pitting limits should be agreed upon by the Engineer, Owner and Contractor before the hydrotest is performed, subject to the criteria of 10.2.1.



CODE Commentary 10.2.xx – Tank foundations shall be monitored and recorded for settlement before, during, and after the hydrotest, as per Chapter 8 (Foundations) of this document. When settlement monitoring exceeds predefined values the design engineer shall be notified immediately.

R10.2.xx - Baseline tank foundation settlement data should, at a minimum, be collected at the following construction milestones: 1) Completion of the base slab, prior to commencement of the

wall construction, 2) Completion of the walls, prior to the commencement of the

roof construction, 3) Completion of the roof construction, 4) Prior to, during, and after the hydrotest, 5) At the start of cool down. Inclinometers, if installed, should be surveyed at the same time as defined above.

10.2.4 - After testing, the inner container shall be thoroughly cleaned, dried, and visually inspected by a qualified person for signs of corrosion, pitting, or degradation. All metal components exposed to the hydrotest water shall be high-pressure spray washed with potable water within 24 hours after completion of the hydrotest. When the hydrotest is performed with brackish or sea water, special care must be taken to ensure that all harmful residues such as Sodium Chloride have been properly removed. All surfaces of the inner tank walls and floor, tested with brackish or sea water, shall be brush scrubbed after the initial high pressure spray wash. A second high pressure rinse with potable water shall be applied after the brushing operation. For concrete tanks, all surfaces of the inner tank walls and floor shall be high-pressure spray washed with potable water within 24 hours after the hydrotest is complete.

10.3 Pressure and Vacuum Testing

R10.3 Pressure and Vacuum Testing

10.3.1 – Pressure Testing Pneumatic testing shall be permitted to be performed concurrently or separately from the hydrotest. Apply an air pressure of 1.25 times the pressure for which the tank vapor space is designed, and hold for 1 hour minimum. Reduce the test pressure to the vapor space design pressure and inspect for leaks at all openings, penetrations, and construction joints. Check the opening pressure of pressure relief valves by increasing pressure



CODE Commentary 10.3.x – Pressure and Vacuum Relief Testing – Proper functioning of all pressure and vacuum relief valves and devices shall be confirmed by:

• Increasing pressure in the vapor space to check pressure relief by increasing pressure in the vapor space.

• Creating a vacuum to check vacuum relief by creating a vacuum with a vacuum pump, or alternatively, by partially withdrawing water from the tank.

Alternatively, In-situ component testing of relief and vacuum breaker valves by applying with test gas applied to the pressure or vacuum to the control valve pressure/vacuum sensing line is permitted. The and set point of controls shall be calibrated with a dead weight tester. Proper functioning of the relief valves and associated sensing systems shall be verified before the tank is placed into operation. The pressure/vacuum in the tank shall be monitored at all times during testing using instruments with alarm settings to guard against pressure/vacuum conditions outside design limits.

R10.3.x – Pressure and Vacuum Relief Testing - During pressure and vacuum relief valve testing, pressure / vacuum levels should be closely monitored for overpressure and excess vacuum. A fail-safe system (e.g., U-tube) should be provided to prevent excessive development of pressure or vacuum. Add to the Commentary R10.3.x - …. Verifying proper functioning of the relief valve system includes, but is not limited to ensuring that the pressure drop in the sensing line does not cause premature reseating of the valves.

10.3.2 - Pumpwells shall be tested in accordance with ASME B31.3, Chapter VI, 345.6 – Hydrostatic-Pneumatic Leak Test. The pumpwells design pressure shall be at least the maximum pump discharge pressure.

10.3.3 - Pneumatic test can be performed concurrently or separately from the hydrotest.

10.4 – Purging into Service

R10.4 - Purging into Service - The provisions of this section apply when the inner (primary) container is concrete or metal.

10.4.1 - Purging shall be a continuous, uninterrupted operation using nitrogen gas with positive pressure maintained within the tank until the start of cooldown.

10.4.2 - A complete purging procedure, including the assignment and duties of operating and supervisory personnel;

• Nitrogen purge gas quality specifications and source of supply

• identification of piping connections • equipment and instrumentation shall be prepared

and approved in advance of the purging operations

R10.4.2 - For purging guidelines, see Reference (35) including references 1, 3, 4, 17 and 19 cited therein. Change reference number. Use references: • Legatos, N. A., and Marchaj, T. J., (1994). “LNG Storage

Tanks – Dewpoint Criteria for Purging into Service”, Proceedings, Gastech 94, Kuala Lumpur, Malaysia.

• American gas Association manual, Purging Principles and Practice, Arlington, VA, 1992.

• Closner, J.J, Corvini, R. H., and Marchaj, T.J., Purging, “Cool-down and Performance Testing of Prestressed Concrete LNG Containers,” AGA Operating Section Proceedings, May 1976, pp. T-210 – T-218.



CODE Commentary • Crawford, D.B., Durr, C.A., and Handman, S.E., “Inside an

LNG Storage Tank,” Gastech 85, November 1985. • Tarakada, R. R., Durr, C. A., and Crawford, D. B.,

“Condensation in the Annulus of a Double-Walled Cryogenic Storage Tank,” Cryogenic Engineering Conference, San Diego, California, August 1981.

• Venendaal, R., “Dryout, Cool-down Keyed Cove Point Commissioning,” Pipeline and Gas Journal, June 1979, pp -38.

10.4.3 - Both the inner tank and the annular Perlite space shall be purged to a final oxygen level of 8% or less by volume, and dried so that all standing water is removed.

R10.4.3 - Prior to the temperature reaching the freezing point, prolonged and excessive drying should be avoided as residual moisture enhances compressive and tensile strength.

10.4.4 - The inner tank shall be purged first, followed by the annular Perlite space.

R10.4.4 - Regardless of the type of the inner container (metal or concrete) the greatest source of free moisture is the Perlite insulation.

10.4.5 - When all samples from all sample points display an oxygen content of 8% or less, the tank shall be sealed and the flow of nitrogen shall be controlled to maintain a positive pressure. Purging shall be considered completed if, 12 hours after the tank is sealed, all samples indicate less than 8.8% of oxygen and dew point temperatures at or below the prescribed levels.

R10.4.5 - Reference (36) recommends that the dew point temperature be maintained at the lower of the following two values:

(a) Below the dry bulb temperature inside the tank; or (b) 14 oF (-10 oC) inside the inner tank, and 50 oF (10 oC)

in the annular space and under the floor. Change reference number. Use reference: Legatos, N. A., and Marchaj, T. J., (1994). “LNG Storage Tanks – Dewpoint Criteria for Purging into Service”, Proceedings, Gastech 94, Kuala Lumpur, Malaysia

10.4.6 - If the entire purging operation is to be accomplished with warm nitrogen, advanced preparations shall be made to begin cooldown immediately after the oxygen and dew point target values are reached.

10.4.7 - Alternatively, purging into service shall be permitted to be accomplished by a combination of warm nitrogen gas followed by liquid nitrogen vapor as prelude to cooldown.

R10.4.7 - If purging is to be accomplished by a combination of warm and cold gas, the following procedure may be considered: Introduce warm nitrogen vapor until the dew point is lowered to 32 oF (0 oC) inside the inner tank, and 50 oF (10 oC) in the annular space; then switch to liquid nitrogen through the cooldown ring. The rate of cooling during this stage should follow the criteria for cooldown so as to avoid excessive temperature gradients in the concrete inner wall. Continue the nitrogen cooling until the target oxygen level is reached.

10.5 – Cooldown

R10.5 – Cool-down - The provisions of this section are intended to address the case when the inner container is concrete. However, provisions also apply to tanks where the inner container is metal, except cool-down rates stated in clauses R10.5.4, and R10.5.7 to R.10.5.9(??) are tank specific and should reference the contract documents.

10.5.1 - A complete cool-down procedure shall be prepared and approved in advance of the cool-down operations, to include:

R10.5.1 - A typical overall cooldown rate is approximately 5.4 oF to 9.0 oF (3 oC to 5 oC) per hour or as stated in the contract documents.



CODE Commentary • the assignment and duties of operating and supervisory

personnel, • Type and source purging gas supply and LNG supply, • identification of piping connections, • equipment and instrumentation, • cool-down abort procedures during initial filling, and

procedures in the event of a long time interval between end of purging and commencement of cool-down,

• Limits on the temperature parameters for cool-down rates of tank components.

10.5.2 - Cooldown operations shall begin immediately after the oxygen and dew point target values have been reached by purging.

10.5.3 - In the period between the end of purging and the start of cooldown, the tank shall be kept at a positive pressure.

R10.5.3 - Before the start of cooldown, the tank should be pressurized to an acceptable internal pressure by adding nitrogen, if necessary. Minimum positive tank pressure not less than 0.7 psi (5 kPa) has been previously used. The pressure should be maintained until cooldown spraying starts. Alternate means for maintaining tank pressure should be provided.

10.5.4 - An adequate supply of pressurizing gas shall be available for maintaining proper tank pressure at all times during cool-down.

R.10.5.4 - Pressurizing gas may be required for this purpose at any time, but in particular for:

• Maintaining tank pressure in the event of an interruption of the cooldown operation, and

• Protection against development of vacuum in the tank during initial cooldown.

Should a continuing decrease in pressure be experienced during initial introduction of LNG, it may become necessary to interrupt the flow of LNG while continuing the supply of pressurizing gas.

10.5.5 - A cooldown spray ring line shall be provided under the suspended deck and equipped with spray nozzles so that cooling down of the tank can be controlled effectively.

R10.5.5 - One of the main objectives of a controlled cool-down for a concrete and metal inner tank is to maintain the temperature gradients across the wall thickness and along the wall height to predetermined levels so as to minimize thermal stresses. Spray shall not impinge on tank wall or be diverted up into the insulated space above the inner tank roof during cool-down or during operation of the tank.

10.5.6 - A network of Resistance Temperature Detectors (RTD’s), thermocouples, or other temperature sensing devices shall be provided along the inside (wet) face, and on the outside face of the inner wall, floor sub floor and annular space to monitor the temperature differences described in R10.5.7 and R10.5.8.

10.5.7 - The cool-down rate as measured at all of the inner tank temperature sensors shall be controlled to within prescribed rates (in degrees per hour). The cool-down rate limits shall be as provided in the contract documents.

R10.5.7 - A typical maximum permitted temperature difference between any two adjacent RTD’s (or thermocouples) is 54 oF (30 oC); and between any non-adjacent RTD’s (or thermocouples) is 90 oF (50 oC), except as limited by the



CODE Commentary recommendations of R10.5.8 below.

10.5.8 - The maximum permitted temperature difference between any two temperature sensors shall be maintained within prescribed ranges. The maximum permitted temperature differences shall be measured between

a. The inner and outer face of the inner wall. b. Any two points along a vertical line on the inside

face of the wall. c. Any two points at the same elevation on the inner

wall face. d. Wall and floor.

The temperature difference limits shall be as provided in the contract documents.

R10.5.8 - Typical permitted temperature differences are as follows:

• Between the inner and outer face of the inner wall: Target 27 oF (15 oC); Maximum 36 oF (20 oC).

• Between any two points along a vertical line on the inside face of the wall, maximum: 5.5 oF per foot (10 oC per meter).

• Between any two points at the same elevation on the inner wall face, maximum: 54 oF (30 oC).

• Between bottom of wall and a point on the floor 15 feet (4.6 meters) away, maximum: 36 oF (20 oC).

10.5.9 - Pressure and temperature readings shall be monitored and controlled continuously to ensure that the limiting thermal gradients, acceptable to the future tank performance defined in Chapter 4, are not exceeded. Maximum pressure differential between the annulus and inner tank shall be no more than 13 mbar with the inner tank always at a higher pressure.

R10.5.9 - For the approximately first three hours of the cool-down, pressure and temperature readings should be made as often as possible on a continuous basis. The maximum time interval between readings should not exceed 15 minutes. During the first 12 hours, rate of LNG flow should be kept below approximately 353 ft3/hr (10m3/hr). After 12 hours, the rate of LNG flow may be increased in increments while monitoring the temperature decrease to avoid excessive thermal gradients. A rate of increase at each 3 to 12 hour interval of approximately 1.32 gal/min (5 l/min) has been successfully used before. This value might be subject to variation depending on the temperature or pressure readings. The estimated maximum flow during the cool-down operation should be approximately 2,300 ft3/hr (65m3/hour). This flow range is only a guideline. The actual flow rate should be governed by the rate of temperature drop between approximately 1.08° – 2.27°F/hour (0.6° – 1.26°C/hour), subject to the recommendations of Paragraph 2 above.. If the allowable rate of temperature change is not within the allowable limits, LNG flow should be stopped until appropriate action is taken to ensure that the cool-down rate is under control.

10.5.10 - The bottom unloading line shall be opened once LNG has begun to accumulate in the tank.

R10.5.10 - When the average bottom slab temperature is approximately -238°F (-150°C) and the thermal gradient in the concrete has begun to stabilize, the flow of LNG through the unloading line to bottom of tank may begin. The tank should not be filled from the bottom unloading line yet.

When the average bottom slab temperature is approximately -256°F (-160°C) and LNG is detected by the tank gauging system, LNG will have begun to accumulate in the tank. At this time the bottom filling line may be opened and the flow through the cooldown line may be stopped.

At this point, before continuing unloading, the emergency shut-down valve should be opened and tested.



CODE Commentary

10.5.11 - The cool-down shall be considered complete when the warmest point on the inside face of the wall is at least -250 oF (-157 oC) and the thermal gradients are within the specified limits.

R10.5.11 - The cool-down can be considered complete if the warmest point on the inside face of the structure is at least -251°F (-157oC). If , despite extended spraying, this condition is not reached, the structure can be considered as cooled quasi complete.

If the cool-down of the structure is complete (so that the warmest point is at least -251°F (-157oC) it can then be filled at specified maximum filling rate of 70,630 ft3/hr (2,000 m3/h), unless the rate of filling is limited by geotechnical considerations (e.g., need for slow-rate loading of foundations, as may be specified in 10.6). The tank should be initially filled with LNG from the bottom fill line to a height of 3m

If the cool-down of the structure is only quasi complete (i.e., some point will remain warmer than -247°F, i.e., -155°C) the rate of bottom filling should be such as to satisfy the criteria outlined in R10.5.9. In this case, readings should be taken at intervals not longer than 1 hour.

10.6 – Settlement and Movement Monitoring

R10.6 – Settlement and Movement Monitoring - The provisions of this section apply when the inner (primary) container is concrete or metal.

10.6.1 - The tank design basis shall provide equipment and instrumentation for the measurement and recording of translational and rotational movement of the inner vessel for use during and after cool down.

10.6.2 - The tank design basis shall include LNG tank tilt settlement and differential settlement monitoring between the LNG tank and external piping to confirm that settlement is within allowable limits as provided in the contract documents.

10.6.3 - Tank foundations shall be monitored in accordance with Clause 8.7 and settlement recorded before, during, and after the hydrotest and first fill of liquefied gases for monitoring of settlement within allowable limits as provided in the contract documents.

10.6.4 - Reference measurements shall be made with appropriate precise instruments to assure that any lateral and vertical movement of the storage tank does not exceed predetermined design tolerances.

10.6.5 - A design professional shall affirm that recorded data for clause 10.6.1 to 10.6.4 parameters are within allowable limits as provided in the contract documents.

10.6.6 - The tank shall be inspected for cold spots where insulation may have formed air pockets in the vertical side walls.



CODE Commentary 10.7 - LNG Tank Fill Methods

R10.7 - LNG Tank Fill Methods - The provisions of this section apply when the inner (primary) container is concrete or metal.

10.7.1 - To avoid tank stratification and to promote mixing in the tank, top and bottom fill nozzles for tank filling of refrigerated liquefied gases are required.

R10.7.1 - The choice of top or bottom fill is based on the density of the liquid currently in the tank as compared to the liquid being added.

10.7.2 - Top fill requires a splash plate on the nozzle outlet to provide a distributed discharge of liquid and removal of entrained vapor. Liquid shall not impinge on walls, splash upward to impact the suspended deck, nor enter the insulated space above or around the inner tank roof.

R10.7.2 – SECTION REMOVED (Duplication)

10.7.3 - Liquefied gas entering the tank shall be flashed into vapor. The pressure differential between the annulus and inner tank shall be monitored and shall be less than 13mbar. The flow rates and temperatures shall be monitored to ensure liquefied gas is flashed before striking the bottom of the tank until the tank bottom has reached a temperature of about 10oC above the liquefied gas boiling point.

R10.7.3 - SECTION REMOVED (duplication)

10.7.4 - Tank pumps shall be used for recirculation if stratification or an unsafe density mixture occurs.


10.7.5 - Top and bottom fill lines shall have in-line flow meters for monitoring of tank fill rates to avoid excess flow and high vibration of fill lines and nozzles during fill operations.




CODE Commentary 10.8 – Decommissioning: Purging out of Service and Warm Up - Two methods of tank warm-up shall be permitted:

a. Natural heat gain through heat exchange through the tank walls and insulation until an equilibrium condition is reached, and

b. Accelerated heat gain by the use of a warm vapor flow into the primary tank, allowing the possibility of reduced warm-up durations.

This section 10.8 addresses the second warm-up scenario.

R10.8 – Decommissioning: Purging out of Service and Warm Up

The provisions of this section 10.8 apply when the inner (primary) container is concrete or metal.

Warm-up of the tank may be necessary due to any of the following circumstances:

• As a prelude to purging and re-entry, such as during decommissioning the tank at the end of its service life,

• The tank is emptied (i.e., LNG level is below the lower limit) and is scheduled to remain empty for 48 hours or more, and

• The cool-down operation is interrupted indefinitely after the coldest spot in the prestressed concrete wall has reached a temperature of -40oF (-40oC) or lower.

The objective of an accelerated (controlled) warm-up is to speed up the warming of the tank by introducing heat mainly by means of warm vapor. Natural warm-up is achieved by natural heat exchange between the tank interior and its environment. In both the accelerated and natural warm-up, the main requirements for the safety of the tank are:

1. To maintain the temperature gradients within the allowable limits, and

2. To prevent over-pressurization of the tank.

If the tank is allowed to warm up by natural heat exchange, it is only necessary to monitor the temperature and pressure readings. If the rate of warm-up is so high as to cause excessive temperature gradients, LNG spray may have to be introduced to slow down the process.

10.8.1 - A complete warm-up procedure shall be prepared and approved in advance of the warm-up operations and shall include:

a. the assignment and duties of operating and supervisory personnel;

b. purging gas quality and source of supply; c. identification of piping connections; and d. equipment and instrumentation

shall be prepared and approved in advance of the warm up operations.

R10.8.1 – A complete warm up procedure should be prepared and approved in advance of the warm up operations and should include:

1) Provisions for continuously maintaining proper operating pressures within the tank while the tank is sealed.

2) Isolation of the tank from all combustible gas lines prior to the introduction of air.

3) Special provisions for the removal of methane vapors from the insulation under the floor, in the annular space and on the suspended deck. For this purpose, the procedures should include provisions for “cycling” gas flow through the purge piping (alternating between supply and withdrawal) - especially through the purge lines terminating in the floor and the annular space.

4) In the case of a nitrogen purge intended to eliminate methane gas prior to the introduction of air the following should be specified:

5) maximum methane content limits to be met prior to the introduction of air, and



CODE Commentary 6) a wait/re-sample procedure.

In the case of preparation for human entry, safety limits on gas content and temperatures, and safety procedures for entry should be established.

10.8.2 - Prior to commencement of warm-up activities, adequate supply of the gasses used during the warm-up procedure shall be present.

R10.8.2 - Prior to commencement of warm-up activities, adequate supply of the following gases should be available: • LNG – LNG should be available for possible spraying from

the cooldown ring if the warm-up must be slowed down because of excessive temperature gradients.

• Methane – Methane vapor should be controlled so that the necessary amount of heated methane vapor will be available for the tank warm-up. As an alternate, nitrogen vapor may also be used for this purpose only if the accelerated warm-up procedure is being used.

• Nitrogen – Nitrogen vapor may be supplied in the event that methane vapor is not available for any reason to complete the warm-up procedure. Nitrogen can be used for this purpose only if the accelerated warm-up procedure is being used.

10.8.3 - Tank LNG liquid level shall be reduced to the minimum possible level using tank pumps.

R10.8.3 - If pumping is not adequate to reduce the liquid level, the under-tank heating system may be used to warm the base slab and allow boil-off of the remaining LNG.

10.8.4 - The exiting gas temperature and inner tank temperature sensors shall be monitored and recorded to avoid developing too high of a warming rate or temperature differentials that may cause undue stress of the tank floor.

R10.8.4 - The tank shall be monitored during warm-up to ensure that the thermal gradients in the tank comply with allowable limits defined in R10.5.4, R10.5.7 and R10.5.8. If the thermal gradients are exceeded during accelerated warm-up, the flow of warm vapor shall cease in order to allow the thermal gradients to return to acceptable levels. If further reduction of thermal gradients is necessary, LNG shall be sprayed into the tank to continue to reduce thermal gradients to an acceptable level. Once the thermal gradients have been controlled, the warm-up operation may restart while continuing to monitor and maintain the maximum temperature gradients as previously indicated.

10.8.5 - If hydrocarbon gas was used for warm up, nitrogen shall be required for removal of flammable gas prior to air being introduced.

10.8.6 - The following variables shall be monitored for the purpose of controlling the purging operation:

a. nitrogen flow, b. internal tank pressure, c. oxygen content, d. dew point temperature, and

temperature of nitrogen purging gas.

10.8.7 - Elevation of Perlite insulation in annulus shall be verified after cool-down and refilled to the proper elevation before placing the tank into service.

R10.8.7 - Compaction of Perlite in the annular space during warm up and settlement of Perlite during cooldown is a common occurrence with LNG tanks.



CODE Commentary 10.9- Record Keeping - For the service life of the component concerned, each operator shall retain appropriate records of the commissioning activities listed in 10.9.1 and 10.9.2.

R10.9 Record Keeping - The provisions of this section apply when the inner (primary) container is concrete or metal.

10.9.1 - Specifications, procedures, and drawings prepared for the LNG tank commissioning activity.

10.9.2 - Results of tests, inspections, and the quality assurance review program.

Outer Tank Wall Inner Tank Wall

Test Pressure

Pressure box with rubber seal to concrete

Braced pressure box back to outer tank wall

Inspection platform

Hydrotest Level


Test Pressure


Braced pressure box back to outer tank wall

Inspection platform

Hydrotest Level

Figure R10.1: General layout of an external pressure box test.



Hydrotest Level

Test Pressure


Braced pressure box back to what?

Inspection platform


Hydrotest Level

Test Pressure


Braced pressure box back to what?

Inspection platform

Outer Tank Wall Inner Tank Wall Figure R10.2: General layout of an internal pressure box test.

Hydrotest Level

Vacuum box with rubber seal to concrete

Vacuum box suspended from roof

Inspection platform


Hydrotest Level

Vacuum box with rubber seal to concrete

Vacuum box suspended from roof

Inspection platform

Outer Tank Wall Inner Tank Wall Figure R10.3: General layout of an internal vacuum box test.

Documents

ACI Committee 376 - Concrete