I have to make time to visit this forum more regularly.

Busy busy busy.


Anyhows, i'm on site looking at the pumps which out team have appraised for nozzle loadings.

I got to thinking, how does caesar attack the issue of nozzle loadings on equipment?

During construction, the fitters normally try to bolt up equipment with no tensions or weights acting on the nozzles, but this is not what caesar is giving in the restraints report.

My fitters asked how they should bolt the pumps up, i replied (as one does) that ensure no strain exists on the equipment before bolt up, ensuring no pull up is used during tightening.

But on the way home, i thought about this and i not 100% happy.

As i see it, to fully satisfy the caesar model, one has to adjust the support/pump locations until the "static" load on the equipment matches that on the restraint report, which would entail using loadcells during construction. not going to happen in west galway!!!!!

Otherwise:
The piping could be: "oversprung" if the equipment is bolted together with no static load on the nozzle
Or the piping could be "undersprung" if the full weight of the pipe is transfered to the nozzle.


I'm a big fan of adding reality into analysis and this has me wondering.
I suspect, that FOS kick in and the problem disappears,


(see attached Pic)

In the "cold" position, if the piping is properly supported, you really shouldn't be imparting a lot of load on the nozzles.

One check that we often run in the model is to "unbolt" the piping from the flange anchor to see what kind of displacements the pipe node undergoes w/o the pump to hold it up. If the pipe goes moving about say, about a 1/2" or more (depending on size, of course as with all of this, judgment comes into play), then your piping system is not adequately supported.

Keep in mind, during this installation/fitup phase, the pipe is empty and the travel stops in your springs had better still be in place. This is the opposite of what Caesar will show you in a SUS case run. So, even though you are seeing loads on the nozzles in the SUS case, that model does not reflect what is going on during installation and alignment phase.

P.S. I'm getting an error when I click on the picture link. Did it get deleted?

A good subject...

Yes, you are correct... reality will not necessarily match what is built.... When in times past I have (unfortunately) produced designs where the "cold" or erection loads would vary significantly from what I knew would be the actual practice, I addressed the differences with an erection procedure.

Do I consider this type of thing all the time? Yes??????????????

Do I write a lot of erection procedures? No!

One must ask themselves is the cost necessary or required. and or justified?


Frankly a lot of things go on that we do not account for and yet things work out! But vigilance is always required.

Sirs.

If the springs are locked in cold load position as available from sustained load output and procured springs are of same stifnesses as used in Caesar-II analysis and we bolt up equipment with no fit-up mismatch in the piping assembly, why will be Caesar II analysis unreal?

But, the problem arises from an effort made in erection to achieve a load distribution what piping stress analyst has not designed.

No fit-up mismatch does not mean absence of tensions or weights acting on the nozzles as even with empty pipe, pipe weight acts and with user springs the situation can be simulated with empty piping sustained condition. One designs the operating condition load at pump nozzle to be minimum, not at the empty sustained load condition.

If I am wrong in understanding, I wish to be corrected by my mentors here.

regards,

sam

"not going to happen in west galway!!!!!"

One of my customers has about 300 load cells and they are using them all the time during pipe erection. Each piece of pipe is entered into CAESAR II with actual WTH, insulation is put on scale... Some owners go to extremes.

Intresting replies.

I try not to use springs because:

A) they are a damn nuisance
B) they are a "band-aid" for poor piping.

I try to "tune" pump supports so that the sustained forces are as low as possible and the pumps pass the force allowables.

But even in bolting up, one is compressing the gasket causing pipe strain.
On Petrochem API pumps, the FOS are good and this is easier, but one should try solving for a EN5199 pump, very difficult.

Superpiper, I've got to disagree. I much prefer a well placed spring support to trying to turn the piping system into a see-saw where the model shows you've got your center of mass about some support so that you've minimized the load on the pump in the model.

To me, that is the essence of a calculator only solution.

If I'm misunderstanding your approach, I apologize. However, that is how I read trying to "tune" the pump supports.

Why is that?

I suppose there are (as always) Pro's and Con's to any solution,
I can think of a dozen for each side of the argument.


Mind you, i only deal in <8" pumps <200c
High temps/creep/vibrations/ etc are not usually encountered.

I'm always open to better ways of doing things..............

My 2 cents worth...

1st Springs are a nuisance....

2cd Springs are very nice for providing a "soft" vertical support that may be tuned to reduce loads at pumps.

3rd Springs take care of "hidden" thermal growth issues so that a fixed +Y doesn't create a large overturning moment on a pump flange for instance. @ 200C for instance a trunnion / staunchion/ dummy pipe will be growing approximately 1/16" /1 - 1.3 ft so lets sy we have a large diameter line with 6" of insulation the staunchions vertical growth may easily be close to a 1/6" taking into account the pipes half diameter and the staunchion. If the line is attached to a cooled pedestal type pum and is a close to the flange then the line may pry up significantly with a fixed support versus a spring.

Not all solutions are simple, neither is this one but a detailed review may lead you one way or another....

There's also a nice phenomenon when you have a FOT reducer attached to a pump. If you put a rigid stanchoin on the larger diameter pipe, there will be a differential vertical growth generated by the offset of the eccentric reducer. This ignores the capacity of the pipe to locally ovalize in the section where the stanchion attaches; another instance where the simple beam theory model with SIF's and flexibility factors falls short of reality.

"Horses for courses" as we say.

I avoid supporting equipment close to the nozzles, (within reason)
it takes some of the flexibillity away from the connection.
Supporting "hot" lines close to the nozzle never looks good to me.

Sometimes a "soft" support is not ideal (two phase flows etc,)

"Not all solutions are simple, neither is this one but a detailed review may lead you one way or another.... "

Agreed, but on lines with Low DeltaT, springs are not always nessesary.
On hot lines as you describe, Sprung supports are considered.

WRT your example,
would not construction tolerance's, Spring creep,misallignments give rise to "deflections" exuivelent to the 1/16th inch growths?

What are difference between WRC 297 & WRC 107, what is the basic criteria for selection of analysis in WRC 297 & WRC 107?

thanks in advance for those who answer my doubet

Use the [Search] function (above the Calendar at the right) to search for "WRC107". You'll find items like this:

http://www.coade.com/ubbthreads/ubbthrea...h=true#Post2912

Besides which what the @%^#@&%&@ do these have to do with pumps please do not muck up the topics!