Hello Every one,
Does anyone could advise me where I can find materials for performing dynamic analysis? and how to input into CAESAR?
we have received a vibration survey from the our client. vibration occured in existing piping piping. I have never had a chance to perform such analysis, I would appreciate if someone could help , recommendation, tips on how to perform such analysis.
what data needed for the analysis?
in fact, we have received a vibration survey showing at certain points the velocities and displacement as well as frequncies...any advice on how to input these values into the system, and shall I consider all, or just input wherever is max? a brief explanation would be appreciable.
thanks in advance

Would this be a good time to talk to your supervisor or employ an expert ?

What you here have is the system answer from a unknow dynamic event.
What you can do is a static run with displacements at the nodes thats move.
The result then is a amplitude.

here no. sr. stress engineer. in my co.
can anybody help me please

panic.

Use displacement and frequency data to determine if system is safe or unacceptable.
Do not use this information as input for CAESAR II - these data are observed responses, not input.

Check the stabilty of the piping system first. Go for MODAL analysis, arrest the lower frequency.

Dear all,

The approach suggested by Ohliger can be used? Please correct me if I have interpreted Ohliger comment wrongly.

The method suggested by Ohliger:
- Put in displacements measured at the nodes that move
- Run the system in static analysis
- Result output: amplitude. How can we get amplitude?

Dave Diehl,

You said do not use displacement and frequency data as input for CAESAR II as they are observed response, not input.

I am a bit confused now.

How do we use the displacement and frequency data to determine if system safe of unacceptable? is it by comparing the displacement output result with measured (observed response)displacement at the site?

Please advice.

By the way, a job well done to the Germans in the semi-final of Fifa World Cup 2014 :-).

Danke/Thank you.

In my opinion...
The observed response is used to confirm that your mathematical model (system stiffness and mass and the applied load and forcing frequency) is adequate in predicting the shape of your vibrating piping system. If they don't "match", you cannot rely on those results.
So now you have a good CAESAR II model. Is there a calculated stress coming out of CAESAR II that can be used to state the system is "safe"? A better question would be - do the piping codes (in CAESAR II) offer an evaluation if these localized cyclic stresses? Basically, No. Your CAESAR II model is a system model that ignores surface detail that serve as very local stress risers or stress concentration factors (SCFs). Do not confuse SCFs with SIFs. SIFs are common in piping analysis and reflect the reduced fatigue strength of the component based on the component shape. And don't get me started on calculated code stress versus actual stress in the system.
There are several documents that focus on de-tuning the system (shifting natural frequencies to avoid resonance with forcing frequencies) rather than evaluating stress levels. There is also a common chart that sets allowable peak-to-peak displacement of a point in the system based on a vibration frequency - this chart defines boundaries such as "Perception", "Design", "Marginal", "Correction" and "Danger". No stress to be seen.
Not a good answer for engineers used to calculating system stresses and comparing them to their code-defined allowable limits.

Dave,

I am more interested in your statement
"Use displacement and frequency data to determine if system is safe or unacceptable."

Many times at site we observed vibration in the piping system and in most of the cases these vibrations are mitigated by providing additional damping or supporting in the piping sysytem.However in few cases they are not controllable by these simple modifications and as a stress engineer at site we have to take decision whether we shut down the system or allow it to be operational.Especially in that case can be use the observed field vibration displacement and vibration frequency as a imposed loading and forcing frequency for harmonic fatigue analysis.
In this case we will reduce/tweak the imposed displacement so that we will get the resonant resposne similar to Field vibartion data . On the same time the fatigue allowable stress from ASME section VIII Div 2 will be reduced for taking the effect of SCF.
I know that thsi can not be called correct dynamic analysis but on a temporary basis can be used this methodology for saying syatem safe or not safe . Please provide your valuable response.

Here is the chart I mentioned earlier. You can find it elsewhere too but this comes from a tutorial presented by Engineering Dynamics Inc. titled Piping Vibration Analysis. The presentation states "these vibration versus frequency criteria can serve as a good starting point in evaluating piping vibrations to screen those systems that need further analyses".

Dave thanks for the chart .However in my case the vibration are exceeding the danger level by 20% i.e. at 2.8 hz max. peak to peak displacement is 2.9 mm. Please provide your comments about the methodology i have given in previous post.

I have little specific advice to give you here.

Dear Mas

Detail survey need to perform at site to assess root cause of vibration. I guess there is excessive vibration at site with high noise.
Piping can create vibration for the following reason
1)expansion bellow is used in the line but not installed correctly with correct type of bellow.
2)Excess flexibility by using spring support to match acceptable pump nozzle load.
3)Excessive load at nozzle due to thermal effect
4)Misalignment of pump nozzle
5)Resonance flequency

Many cases it is found the piping is not the real culprit for pump vibration
1)Pump base need to have enough rigidity
2)Misalighment is motor/pump & bearing
3)Check suction strainer for any possible blockage

Regards
Habib