St Llc 0702 Pres

6/3/2013

1

Presentation Title:

ASME ST LLC 07-02

Recommended Changes to Appendix D

Presentation Summary -

Appendix D of the B31 codes offers stress intensification factors (SIFs) for

common fittings in piping systems.

It has been known since 1987 that a number of these SIFs are incorrect

and give incomplete guidance.

The omission of k-factors for branch connections in Appendix D is also a

known source of potential error in piping system analysis.

ASME ST LLC 07-02 provides more applicable data to address these

issues.

Presentation Contents:

1-Introduction of B31.3 Appendix D

2-Identification of parts of Appendix D where SIFs and k-factors might be affected

3-Schneider and initiation of WRC 329 in mid 1980’s

4-WRC 329 corrections for Appendix D SIFs and k-factors for branch connections

5-References and work after WRC 329

6-Introduction of ST-LLC 07-02

7-Examples in ST-LLC 07-02

8-Additional Examples

6/3/2013

2

B31.3 Appendix D

6/3/2013

3

6/3/2013

4

Piping Code Limitations:

• D/T > 100 (not a limit for FEA)

• 0.5 < d/D < 1.0 may be non-conservative. (not a problem for FEA)

7

Some additional notes in the 2010 version of B31.3 that contrast

a little with what’s in ASME ST-LLC 07-02.

6/3/2013

5

There were concerns in the 1980’s from R.W. Schneider

that Code SIFs for branch connections were not

conservative when 0.5 > d/D < 1.0.

WRC 329 was intended to address this problem, but E.C.

Rodabaugh found a number of issues with the present

Codes. Most of those issues have been addressed with

the technology of the time for ASME III NC/ND, but

have not been addressed in the B31 Codes.

WRC 329

6/3/2013

6

6/3/2013

7

p.22

6/3/2013

8

“Silly” Overestimate of the Stress:

ST-LLC 07-02 8.9767/1.0439 = 8.599 Times TOO Big

ASME Section III 8.9767 / 2.1 = 4.27 Times TOO Big

FEA SIF 8.9767 / 1.0 = 8.9767 Times TOO Big

6/3/2013

9

6/3/2013

10

6/3/2013

11

22

WRC 436

iir(test) = 1.050

6/3/2013

12

Basically – don’t use the effective section modulus with the

new rules. It is easier for the rules to provide a more

appropriate i-factor..

AND – Fix Weldon Fittings (Olets)

6/3/2013

13

Excerpt from Table 1, p.4.

WRC 329 – pp.31,32.

Reformatted WRC 329 Table 13 for Mob Evaluation (10 thru 17 – Olets)

6/3/2013

14

Mob comparison of test, B31.3 and ST-

LLC 07-02 for UFTs, EXTs, and Olets.WRC 329 – Fig. 16

…

6/3/2013

15

After WRC 329 (after 1987)

Since WRC 329 in 1987 there have been a number of

papers published that can affect Appendix D guidance:

2003

1999

2001

2007

1998

6/3/2013

16

2005

2006

1993

2002

1998

2004

1998

2001

1998

1997

2008

6/3/2013

17

2002

1989

1994

2006

1994

Khan’s tests were run in Oklahoma

on the WFI/PRG Markl Test

machine that is presently in PRG’s

offices in Houston Texas.

These tests were run to supplement

Markl’s test so that some voids in

the test parameter ranges were

addressed.

6/3/2013

18

Markl Fatigue Test Machine Setup for k-factor Test

Blair tests run

below 40Hz.

6/3/2013

19

Item Markl

Test

Hinnant

Test

ST-LLC

07-02

UFT

in-plane

3.74 2.57 2.89

Pad

in-plane

2.91 1.85 2.08

Blair SIF Test Summary (Avg Values)

Item Test ST-LLC

07-02

WRC 497 Wais

UFT

in-plane

4.422 5.89 6.86 4.196

Pad

in-plane

2.062 2.183 4.034

Blair k-factor Test Summary (Avg Values)

First recorded fatigue tests for piping recorded

by Markl. Dynamic shaker tests conducted by

Blaire between 1935 to 1945.

PRG k-factor, i-factor and I-factor tests

conducted in 2013 – Houston Tx.

Piping Fatigue Tests from 1935 to 2013

6/3/2013

20

What do the newer tests and

documents indicate?

6/3/2013

21

42

What are the errors – why didn’t I see them in all my years working in pipe stress?

6/3/2013

22

pp. 36 & 37.

44

6/3/2013

23

PRG regression matches Wais/Rodabaughsurface from EPRI 110997 while PRG FEA data matches Widera from WRC 497.

6/3/2013

24

Some Possible New SIF

Equation Forms

?

?

6/3/2013

25

49

Nothing should be weaker than a fabricated tee (with equal dimensions),

or more flexible than a fabricated tee (with equal dimensions),

No other contoured fitting can be stronger than a welding tee.

i(ir) ≥ i(or) and i(ob) ≥ i(ib)

Relationship between UFT and

WLT in ASME III NC/ND:

The t/T Problem

6/3/2013

26

6/3/2013

27

53

t/T=0.3

t/T=1

t/T=2

Welding Tees

6/3/2013

28

56

6/3/2013

29

57

WPW-Welded

WPS-Seamless

58

6/3/2013

30

59

How can welding tees be updated?

1-Look at existing Code.

2-Recognize that welding tees always have a lower SIF than UFTs

3-Look at UFT variation.

4-Ignore 0.5 < d/D < 1.0 problem

5-Run FEA models with smooth contours and local thicknesses

6-Recognize that i-factors and k-factors increase and decrease together but not

proportinnally. Follow guidance for “reasonableness”. Tees aren’t as thick as they

used to be 1.5-to-3.0 times more than nominal wall. Generally now it’s 1.1 to 1.5

times with no crotch thickening.

6/3/2013

31

Olets

For fittings with larger d/D ratios the bevel below the parting line in the circumferential plane may reduce the fitting

thickness available for pressure containment in the circumferential plane and in some cases may reduce it such that it

is equal to or smaller than the wall thickness of the attached matching pipe. This is not likely the intent of the

manufacturer however. These dimensions are not controlled, and a number of the fittings are produced by copying

those of another manufacturer, which may result in angles from one fitting being applied incorrectly to another.

The amount of reduction is a function of the dimensions selected by the manufacturer. These dimensions are not

controlled by any standard. Welded outlet fittings are manufactured to typically MSS-SP-97 and in some part to

B16.9.

As the d/D ratio gets smaller the thickness reduction also gets smaller and the bevel shown below the parting line in

the figure above vanishes. In the smaller d/D case, below the parting line, the welded on fitting profile is similar, if

not identical to that of a typical thickened straight nozzle body. The dimension (di) is equal to the inside diameter of

the matching pipe, and the outside diameter, (while not controlled by any standard), is often approximately the

dimension X in Table 14 of B16.5, or given by the relation included in Note 12 of the 07-02 project.

For welded outlet fittings, it is expected that the i-factors and k-factors converge to those for Sketch 2.3 with a

thickened, straight barrel length as described in Appendix B Fig. 5 as d/D < 0.5.

6/3/2013

32

63

64

6/3/2013

33

Modeling Branch

Connections

6/3/2013

34

67

6/3/2013

35

69

70

6/3/2013

36

Testing

72

6/3/2013

37

73

74

6/3/2013

38

6/3/2013

39

77

Analysis

6/3/2013

40

79

The NB3686 model with a rigid element from the center to

the surface of the pipe is used for all d/D ratios?

How it’s Tested How it’s Used

6/3/2013

41

Reporting

Comparison to current Codes and MethodsComparison to Test Data82

6/3/2013

42

83

84

As part of the project all i-factors and k-factors through 2007 were collected.

6/3/2013

43

85

86

6/3/2013

44

87

WRC 166 – McClure

(Similar to Blair Tests?)

88

6/3/2013

45

89

72 of these comparisons are provided. For each branch connection sketch

there are plots for: iib, iob, itb, iir, ior, itr, kib, kob, ktb, kir, kor, ktr

6/3/2013

46

6/3/2013

47

6/3/2013

48

95

Major Features of Aligned Equations

1) Flexibility Factors Added for all Branch Connection Components

a) k-factor for 90 deg. elbow adjusted per N-219 and Fig. 19 in ORNL/Sub-2913/7.b) Run and Branch Flexibilities Provided and Examples Presentedc) Branch Connection Flexibilities Modified for Attached Flanges

2) SIF Updates

a) Out-of-plane d/D Inflection Includedb) t/T Inflection Includedc) Reduction of SIFs for Run where Appropriated) Weld-on Fitting SIF Correctione) In-plane, Out-plane and Torsional SIFs Providedf) Locally Thickened Branch Rules Clarifiedg) EPRI Rodabaugh/Wais Results for Concentric and Eccentric Reducers Addedh) Branch Connection Figure Clarification

3)Corrections, Note Changes, etc. Recommended in WRC 329 Implemented

ST-LLC 07-02 is three documents:

�Replacement for Appendix D

�Example Problems illustrating application with discussion

�Detailed calculation example

6/3/2013

49

Replacement for Appendix D:

Note that for contoured fittings (B16.9 tees, extruded tees, etc.) the k- and i-

factors are proportional to d/D raised to a constant.

6/3/2013

50

Note that for fabricated fittings, e.g. pads, UFTs, olets, i- and k-factors are a cubic

function of d/D. These d/D functions produce a “hump” when 0.5 > d/D < 1.

6/3/2013

51

WRC 329 Para 4.2.4, “A puzzling aspect fo the Table 6 [extruded tee] results is that

the [i-factors] are higher than would be predicted by Code equations for UFTs.”

(See next slide for geometry discussion.)

6/3/2013

52

103

Difference between the middle Sketch 2.3 and the first Sketch 2.5

6/3/2013

53

From Note 15.

6/3/2013

54

6/3/2013

55

6/3/2013

56

111

6/3/2013

57

114

Reference for Note 10 Guidance for Laterals

6/3/2013

58

6/3/2013

59

6/3/2013

60

6/3/2013

61

6/3/2013

62

6/3/2013

63

Without considering the branch connection flexibility of

the 12x30” fabricated tee at point 15 the out-of-plane (Z)

bending moment at point 15 is 372,000 in.lb. Including

the branch connection flexibility reduces the bending

moment to 41,832 in.lb., a reduction of 8.8.

6/3/2013

64

128

Example No. 5 Heater Piping

6/3/2013

65

Fails without k-factors,

passes easily with k-factors.

Studying the system, there

are 6 welding tees in line

with very little straight pipe.

The welding tees control the

planar stiffness of the model.

(With k-factors stress is

reduced by 70%.)

Moment increase by

more than two times

at vessel nozzle when

flexibilities are

included.

The unbalanced load in a typical

variable support spring design,

causes more cold load movement.