Flexibility and Seismic analysis - Piping system 2

[GD2]

Experts,
Many of you do flexibility analysis of a piping system (say ASME B31.3). Has anyone experience in doing a seismic analysis of the same piping system?
There could be instances when you have done the flexibility analysis nut a third party contractor does the seismic analysis. In this situation, how do you analyze and implement both reports? Do you rerun the flexibility analysis once you implement the seismic analysis?

GDD
Canada

 

[sentrifice]

They should be done as part of the same calculations. The stress analysis should calculate stress and deflection for all cases applicable to your system. So if your piping gets hot and is in a seismic area, then the load combinations should reflect that.

I'm still an amateur compared to most here, but as far as I'm aware the only practical way to do this is for one team to do the complete stress analysis which would encompass "flexibility" and "seismic" analysis. Also, any reputable stress analysis software is going to have seismic provisions so it wouldn't make sense to do it separate.

 

[Snickster]

Seismic along with wind loads are considered Occasional Loads by ASME B31.3. A stress analysis for piping includes these loads along with thermal loads, internal pressure loads, etc. There are no separate calculations for seismic loads by other 3rd party design firms. The design engineering company is responsible for all stress analysis.

Input screen worksheets are available in pipe stress programs such as Caesar II, to input wind and seismic loads on the piping and FOR building the overall piping system model. After the piping system is modeled and load defined for each segment of piping, the program is run by next defining the combination load cases.

For instance, ASME B31.3 requires 3 basic load cases (load combinations) to be analyzed and each load case has its own allowable stress limit:

Sustained Loads SUS
Thermal Displacement Loads EXP
Occasional Loads OCC

Operating Load OPE cases and Hydrostatic Load HYD cases are also performed to get maximum loads such as on equipment nozzles and pipe supports, but do not have to meet any allowable condition per ASME B31.3 as long as the 3 basic code stresses indicate above are within the allowable stress limits.

Typical list of load combinations to check for are:

Where
W=Weight of piping and contents
P=Internal design pressure
T=Design temperature
H=Hanger loads for systems with spring hanger supports
WW=Weight during hydrotest
HP=Hydrostatic test pressure
WIN=Wind Load
U=Seismic Load
Lx=Individual load combination case
Lx-Ly=One load combination is subtracted by another load combination to get a single isolated load effect such as wind and seismic, or thermal expansion.

Basis load cases
L1 W HGR (if spring required)
L2 W+T1+P1 HGR (if spring required)
L3 WW+HP+H HYD
L4 W+T1+P1+H OPE
L5 W+T2+P1+H OPE
L6 W+T3+P1+H OPE
L7 W+P1+H SUS
L8 L4-L7 EXP
L9 L5-L7 EXP
L10 L6-L7 EXP
L11 L4-L5 EXP (Total stress range)


Basis load cases + Wind/Earthquake
L1 W HGR
L2 W+T1+P1 HGR
L3 WW+HP+H HYD
L4 W+T1+P1+H OPE
L5 W+T2+P1+H OPE
L6 W+T3+P1+H OPE
L7 W+T2+P1+H+WIN1 OPE
L8 W+T2+P1+H+WIN2 OPE
L9 W+T2+P1+H+WIN3 OPE
L10 W+T2+P1+H+WIN4 OPE
L11 W+T2+P1+H+U1+U2 OPE
L12 W+T2+P1+H-U1+U2 OPE
L13 W+T2+P1+H-U1-U2 OPE
L14 W+T2+P1+H+U1-U2 OPE
L15 W+T2+P1+H+U2+U3 OPE
L16 W+T2+P1+H-U2+U3 OPE
L17 W+T2+P1+H-U2-U3 OPE
L18 W+T2+P1+H+U2-U3 OPE
L19 W+P1+H SUS primary stresses for check
L20 L7-L5 OCC suppress wind/earthquake only
L21 L8-L5 OCC
L22 L9-L5 OCC
L23 L10-L5 OCC
L24 L11-L5 OCC
L25 L12-L5 OCC
L26 L13-L5 OCC
L27 L14-L5 OCC
L28 L15-L5 OCC
L29 L16-L5 OCC
L30 L17-L5 OCC
L31 L18-L5 OCC
L32 L20+L19 OCC primary stresses + wind/earthquake for check
L33 L21+L19 OCC put Comb Method on Scalar for + only
L34 L22+L19 OCC
L35 L23+L19 OCC
L36 L24+L19 OCC
L37 L25+L19 OCC
L38 L26+L19 OCC
L39 L27+L19 OCC
L40 L28+L19 OCC
L41 L29+L19 OCC
L42 L30+L19 OCC
L43 L31+L19 OCC
L44 L4-L19 EXP secondary stresses for check
L45 L5-L19 EXP
L46 L6-L19 EXP
L47 L4-L6 EXP total stress range for check

 

[RVAmeche]

Note "flexibility analysis" is an out of date term that primarily applies to the old school method of looking at charts and roughly calculated displacements and making sure the piping is flexible enough.

Modern stress analysis is very different, accounting for friction of supports, non-linear supports, support lift offs, varying temperatures/conditions, etc. As others have mentioned the stress analysis should include seismic if it's required.

 

[KevinNZ]

Defiantly one stress model for design/operating and occasional loads.

Often an external expert is used to define the seismic actions for a project, but they do need to understand piping design so they can advise the appropriate seismic design factors.

B31E is worth a read.

 

[GD2]

Typically Caesar II and Autopipe are very common for stress analysis including Occasional load analysis.
My question is does these programs takes into account the G-Loads( or g-factors) in 3-directions(East-West, North-South and Vertical)from the local building Codes or IBC, if it is not defined by the Project/User?

GDD
Canada

 

[KevinNZ]

Autopipe has Spectrums for IBC 2006, Euro EC8, EC8 IS 1893-2002. All directions covered.

but you need to know what you are doing to run spectrum analysis.

 

[XL83NL]

My question is does these programs takes into account the G-Loads( or g-factors) in 3-directions(East-West, North-South and Vertical)from the local building Codes or IBC, if it is not defined by the Project/User?

Not sure if this answers your questions (or if I understand you correctly), but AutoPIPE takes the input for seismic that you provide it with, regardless who supplied you with that input.
For ASCE 7, you can enter the ZIP/Postal code (provided the location your pipe stress model will be situated is in the US of A). But then you still need to provide with additional parameters for site class, Ip, rp, ap, Z, etc.

Huub
- You never get what you expect, you only get what you inspect.

 

[RVAmeche]

If you're running stress analysis you need to know the Code requirements and the programs. It's not something where you can just say you followed Caesar II's static seismic wizard and it did whatever, hope its right. That's not how it works.

This is doubly true for full spectrum response analysis and capturing things like varying building/structure heights, etc.