i have a question on Spectrum analysis:
i want to calculate the steam hammer in power plant pipe,and the dynamic analysis was conducted on PIPENET .the"*.DLF" files were created by caesar through "*.FRC" file created by pipenet.
so the questions are:
when "spectrum load cases"is invoked,at the first column,spectrum name is specified separately,here, i have 30 spectrum files,so 30 separated spectrums are specified(see pic02),and,the combination of the 30 spectrum is specified in the 31st load case(see pic01),
and next i invoke the "static/Dynamic combinations"(pic03),to combine a "sus" load with each of the spectrum load cases(pic03)for stress check,this is the first case.and in the"static/Dynamic combinations"(pic04) a sus combine the above specifed combination case D31. this is the second situation.

after calculation,the "sus+D1+D2+...+D30" and the"sus+D31" differs from each other.which method is right
D31 is the combination of D1,D2,...D30(in the"spectrum load cases"window)

thank you very much~~

[img]http://phrenzxh.blog.163.com/editPhoto.d...080074082086095[/img]
[img]http://phrenzxh.blog.163.com/editPhoto.d...080074082086095[/img]
[img]http://phrenzxh.blog.163.com/editPhoto.d...&albumId=-1[/img]
[img]http://phrenzxh.blog.163.com/editPhoto.d...080074082086095[/img]
[img]http://phrenzxh.blog.163.com/editPhoto.d...080074082086095[/img]
http://phrenzxh.blog.163.com/editPhoto.d...080074082086095

I won't tell you which one is right but I will offer an explanation as to why they are different.

Your case D31 combines several shocks that have their own combination rules. The Control Parameters list show these rules - Directional Combination Method (default is absolute) and Spatial Combination Method (default is SRSS). "Directional" controls the summation of shocks defined in the same direction and "Spatial" controls the summation of the different direction shocks.

Your Static/Dynamic Combinations have their own combination directive. This setting is found on the Combination window. The settings are absolute and SRSS. I believe the default is absolute.

I suggest you compare your numbers after rerunning the analysis with the same settings throughout.

= = = = =

A separate matter...

The force response spectrum approach that you are using ignores the timing of each "hit" in your system. Force response spectrum analysis is good for looking at a single impulse but I would point you to time history analysis if you wish to see these 30 events working one after the next, together in time.

thank you very much for your sugesstion,i would try"time histroy" analysis on steam hammer.
another question, when analyzing steam hammer in steam pipeline ,what method is commonly used in USA or other countries,i am from China.thanks~

Here is my opinion...
We have great analysis tools today to calculate all sorts of loads and ressponse. From your transient hydraulic analyis you can generate time histories to represent the hammer load for every leg in your system. And you can push that data through CAESAR II and wade through the results to evaluate your design. Pure number crunching.
But engineering is the practical application of the science using experience. Many hammer issues are addressed, not with analysis, but with construction and operation guidelines. If analysis is required, a more general, wide-ranging approach would be appropriate when specific time history data varies. So a static equivalent load applied to the longer runs might do a better job in bracketing all the possibilities and confirm a more reliable design.

Hello Zhao,

Your piping model is a good candidate for Time History analysis.
But remember, PIPENET interface creates water hammer dynamic input for CAESAR II automatically. So, you shouldn't create dynamic load cases. Only create DLF files then in CAESAR II dynamic you should create spectrum/force set/load cases manually with Time History analysis type. If you open DLF files with Notepad, generated by PIPENET interface, you'll see max. force magnitude. Be careful, DLF files contains Time vs. Force Multiplier.

Regards,
Farhad