B31.1 cyclic condition question 1

[JAYDEE23]

I am looking for some direction with B31.1 cyclic condition. We make pipe components (flow measurement venturi). Our meter is just installed in the line and is more or less a small "pipe" section. Until now i have not dealt with cyclic condition. The current component i am working on specifies "designed for cyclic service (N<7000) per B31.1"

As i look into B31.1 it seems I need much more info than the standard temp, pressure, fluid, line metal, etc that ive been provided, to consider cyclic in my design.

I assume the piping run itself has been designed for cyclic. I know this is an issue between me and the customer but any insight into cyclic would be appreciated...

 

[davefitz]

The best primers for understanding and addressing cyclic fatigue effects in piping systems that I have found are the european pressure equipment directive PED code EN 12592-3 ( boiler fatigue damage- availalbe as an ANSI code)and the older code TRD 301 annex 1 ( german boiler code). These provide methods for estimating thermal stresses in simple components ( due to cyclically fluctuating pressure and temperatures ) and accurate means of indentifying and counting fatigue cycles and estimating fatigue damage. Theyu do not include pipeline flexibility analysis or creep fatigue interaction; you need other programs to inlcude tose affects .

In the case of piping components, such as a flow element or valves welded to a pipe, the peak thermal stress is likely at the weld interface between the thinner pipe and the thicker valve. The magnitude of this stress is related to the rate of change of fluid temperature , the wall thickness ( squared) of the valve, the valve alloy's thermal diffusivity, and the estimated stress concentration factor at the weld- this a related to the post weld inspection procedure and the rate of chagne of wall thickness in the transition from thin pipe to thick valve. The worst case would be a fast thermal transient at a thick valve welded to thin pipe with a sharp angle at the chagne in wall thickness , and the weld assumed to be un-inspected at the ID .

The client will need to tell you the pipe wall thickness at the weld interface, his expected post weld inspection procedure, the rate of change of fluid temperature ( during cold startups) , and the number of these transients to be expected over the life of the piping system. You may need to counter by requesting teh clinet perform an inspection of ID after welding, and to use a transition piece at teh weld interface to minimize the gradient in wall thickness so as to allow the use of a lower stress concentration factor.

 

[JAYDEE23]

First thank you for the response. As it stands i am awaiting a revised spec removing our responsibility for cyclic. We are contracted for the component only. All aspects and design of the pipe line, including installation of the meter, are not our responsibility.

FYI the condition of the weld joint is same material, same wall thickness ID & OD, and full pen. V groove...

 

[ApC2Kp]

JAYDEE23,

The B31 allowable stesses for the materials in Table A-1 thru A-7 have been established for 7,000 cyclces based on a daily start / shutdown for approximately 20 years. If there is more frequent thermal or pressure cycling, then the designer would need to evaluate and apply factors to the basic allowable stresses to de-rate for extended cycle life in those high cycle applications > 7,000 cycles.

Most pipe fittings are designed for pressure capability that is equal to straight pipe of compatible size. The pipe elbow or tee of say nps-2" size would have pressure design equal or greater than the nps-2 pipe. The exception is for flanges which have design pressure and dimensions to ASME B16.5 pressure classes. Your flow venturi might be designed to the ASME B16.5 pressure classes for wide application into piping with standard flanges.

 

[JAYDEE23]

The meter construction is a weld in type (no flanges). This considered I am limited with the facts
-I must match existing pipe being welded to
-I can only calculate with what i am supplied..

I was confused on why the customer would ask me to design >7000 cycles and not supply any additional info. other than temp, pressure, fluid, material...

It looks like it is a copy and paste error with this customer...arrgg...

 

[JohnBreen]

ApC2Kp

Please excuse a little "nit picking" here.

"The B31 allowable stesses for the materials in Table A-1 thru A-7 have been established for 7,000 cyclces"

Well, No. The Allowable stress at temperature, as found in the cited Appendices, are based up 1/3 yield or 2/3 ultimate strength, independent of cycles. The allowable stress RANGE equation (B31.1, paragraph 102.3.2(c)), includes an allowable stress range reduction factor, f, that is based upon 7,000 cycles being 1.0 (no alllowable stress RANGE reduction).

"Most pipe fittings are designed for pressure capability that is equal to straight pipe of compatible size."

.........and schedule. We are talking ASME B16.9 FITTINGS here. And ALL, not some, are base upon burst test pressures being the same as straight pipe of the same size and schedule.

"The exception is for flanges which have design pressure and dimensions to ASME B16.5 pressure classes."

NOT an exception really as we are talking B16.5 FLANGES here - apples and oranges.

JAYDEE23

Yes, the pressure boundary your supply with your equipment must be designed to the same design data as the piping system in which it is to be installed - you do not want your equipment to be "the limiting component".

Yes, you client will have to give you ALL the design information including the expected FULL cycles and the expected PARTIAL cycles. There will be far fewer full cycles with a temperature range from the coldest design temperature (shut down on a winter night) to the hottest design temperature. There will be many mopre partial cycles from varying ambient temperatures to operating temperature.

You might also want to look at the ASME B31 Code for Pressure Piping, B31.3, Process Piping as we have modified the section addressing thermal cycles to include "F" factors greater than 1.0 for cylcles less than 7,000. Generally, we keep B31.1 rule more conservative than B31.3 rules.

Regards, John.

 

[JAYDEE23]

Thanks for feedback...appreciated...

 

[stanweld]

B31.1 design allowables are based on a 3.5:1 safety factor; B31.3 design allowables are based on 3:1 safety factor - based on minimum specified tensile strength:design allowable stress. Or 2/3 specified minimum yield.

The basis for B31.1 "...are the same as those in the ASME Pressure Vessel Code, Section II, Part D, Appendix 1..." ref. 102.3.1 (C).

Fatigue (Cyclic) service is not considered in ASME II, D, Appendix 1.