Hi all,

According with topic: http://65.57.255.42/ubbthreads/ubbthread...=true#Post44919 ,

"The difference between the large ID of the bellows and the small ID of the bellows will provide an area upon which the pessure will develop a compressive load on the nozzle."

So,

It's the same method that we can find at EJMA Standards. If we do a hand calculation (according with EJMA), it's possible to isolate the compressive load on the nozzle and the load that goes to pump anchor.

But how I can simulate these loads isolated in CAESAR II? And for API 610, what is the load used for calculations? Load on pump nozzle or Load on pump anchor?

For a simple example, please use Example 06 encountered in EJMA Standards Appendix J, page J-19.

Thanks in advance,

Miyamoto

Good morning,

Follow example attached.

Regards,

Miyamoto

Dear all,

Sorry for up this thread again, but the doubt persists.

Someone can help me?

Thanks again,

Miyamoto

Dear all,

Sorry for up this thread again, but the doubt persists.

Someone can help me?

Thanks again,

Miyamoto

Maybe you can use W+P1+T1+F1 where F1 is your Fp. But.....

You are going to transfer a thrust force on the pump case and then on the baseplate which is not usually taken into account by the pump vendor. So it will have less importance that you comply with API 610, your pump will not withstand this thrust.

Regards,

Hi Miyamoto

Pump nozzle will experience less load. Pump anchor will experience more force & moment due to thrust load.
So for large bore pipe with high pressure it is not advisable to use untied bellow. For tied bellow tie rod will absorb the thrust load.

Regards

Habib

Thanks danb and shr for these explanations.

But, according to EJMA these loads acting on pump nozzle is less than acting on pump anchor. So, how can I simulate in CAESAR II the load that goes only to nozzle (for API 610, is this important load?).

If we model a pump nozzle as a restraint ANC, the load resultant is thrust load, and if we use this force to compare with API 610 allowables, we will commit a mistake due API 610 verify only loads on the nozzle.

danb, but if the load acting in pump casing is bigger than nozzle, this load will be response for shaft misalignment, am I right?

So, I have 2 options:

1) I do hand calculations as shown in EJMA for pump nozzle loads and use analysis module of CAESAR with these loads.

2) Simulate only the load on pump nozzle (desconsidering as ANC)

What I need to do?

Regards,

Miyamoto

If you want an API 610 check, fine, you can do manually or you can apply a force in opposition with your thrust in order to emulate the effective load on the nozzle. Alternatively you can do as follows. Do not fill in the “Effective ID” field then calculate the real force on the nozzle and insert it in the “Force/moments” field.
However you will be responsible for the fact that you apply a thrust on the pump case. If the pump vendor is informed and accept this, fine. If not, do not use untied bellows.
Note the force in Anchor A from your example.

Hope I've made myself clear.

Regards,

danb,

Ok. My last doubt. What matter for vendor or API 610 check? Effective load on the nozzle as show EJMA Example? Or thrust load applied on pump case/anchor?

Only for understand: API and vendor don't evaluate force on case/anchor? It means don't matter if nozzle checked by API was ok and load on pump case is overloaded? If this is all right, how can I eliminate thrust forces in a straight pipe considering a set of pumps in series(begin as discharge and end of pipe as suction of another pump)?

Thanks again,

Miyamoto

Vendor do not know that you are going to use untied bellows unless you inform him.

Hi all again,

Sorry for ressurrect an old thread, but I have doubts about this subject yet.

Using an untied bellow, according to EJMA, the load aplied at pump nozzle is different of the load that aplied at pump anchor (this is the thrust force). And API 610 Standard, Table 5 show allowables to compare with the model. What load I use to compare to API 610? Load at nozzle or load at pump anchor?

Best Regards,

(And sorry again)

Miyamoto

Please seniors, help me with this subject.

What does API610 tells you?
Table 5 - Nozzle loadings!

Hi Miyamoto,
In my opinion you should not use the untied bellow inline with pump nozzle. The reason is the thrust force; it produces higher loads on your nozzle. API-610 tell you to check the nozzle loads for pumps. It will be great help if you can inlcude your geometry (pipe route) in your reply. Then only solutions will be more practical.

Regards,
R.K.

DEAR SIRS,

I HAVE A QUESTION ABOUT THIS SUBJECT. WHY DOES NOT PRESSURE THRUST ACT ON PUMP FOUNDATION BY USING OF A TIED EXPANSION JOINT?(BUT P*ID(PIPE INNER DIAMETER) EXISTS IN TIED AND UNTIED EJ)
COULD YOU ATTACH A SCHEMATIC FIGURE FOR A CLEAR ILLUSTRATION?

Hi Miyamoto,
There is a pressure thrust exist in tied expansion joint also, but that presusre thrust is taken care by the tie rods, so generally it woun't have any effect on your calcualtion.

Tie Rods: - are the devices which usually in the form of rods or bars, attached to the expansion joint assembly whose primary function is to continuously restrain the full pressure thrust during normal operation while permitting only lateral deflection. Angular rotation can be accommodated only if two tie rods are used and located 90 degrees opposed to the direction of rotation. I hope this will help you!!

Regards,
R.K.

For piping systems that contain no expansion joints, the axial loads include a longitudinal force calculated as Pressure X Internal Pipe Area.
Reporting this force to the pipe cross-sectional area, you obtain the stress due to pressure load, usually counted as "PD/4t".

In "EJ with no rods" case or in case the rods are not working, the EJ "breaks" the piping system in two parts and the pressure thrust is acting on both of them.

I suggest you to see also
http://65.57.255.42/ubbthreads/ubbthread...=true#Post33752

You can find there a sketch for Peng's book.
In fact, there is an warning that the existence of roads is not always a guarantee. As you can see, Mr. Peng made a suggestion on the piping part.

Best regads.

Obviously, I must say "the existence of rods is not always a guarantee"...

Mr. Peng point is that the pressure thrust will move the piping on the sliding support, reestablishing the "piping continuity" thru "tie-rods".
Of course, it is a good idea.

However, in my understanding, when there is a pressure thrust moving the piping (as in book's sketch), it means that there is also a coincidental pressure thrust acting on Pump side. Of course, on that side the force is not moving the pump...however, theoretically, this short-time force is anyway acting on pump. But is only a short-time force, until the rods are working again; in addition this "process" is smooth enough.

Just my opinion, of course.

DEAR MARIOG,


WHEN SHOULD I CLOSE THE NUTS OF THE RODS AND WHEN IT SHOULD HAVE A GAP?

The discussion in Peng’s book assumes using an EJ to solve the equipment load problem due to piping thermal expansion.

Mr. Peng criticism is addressed to the fact "The anchor is so important in the application of bellow expansion joints that many engineers seem to automatically place an anchor or anchors in an expansion joint installation."
Because the anchor point does not move, the pipe located between the anchor and equipment expands into the joint. This leaves the nuts loose with gaps.

So you have now other problem: "when tie-rods become loose, they no longer resist the pressure thrust force."

Mr. Peng conclusion is "The correct installation is to allow the support point to slide as shown. The movement of the elbow ensures the tightness of the tie-rods, because the pressure keeps pushing the elbow until the rods are tight. The tightened tie-rods can then absorb the bellow pressure thrust force, leaving no unbalanced pressure thrust force on the machine."

My comment was that things are not so simple, because-theoretically- "the pressure keeps pushing the elbow" is coincident with "the pressure keeps pushing the equipment casing". When you have an anchor attached to elbow, this fact is evident and you have a long-time negative effect of pressure.
In my understanding, when you have an sliding support there, this effect still exists but is likely to be a short-time state if the pressure trust can easily move the piping on sliding support. In this case the developed nuts' "gap" would be quite small and smoothly compensated by the pressure trust force.
But the real behavior (and actual gap) depends on pressure thrust magnitude, friction effects and stiffness of the piping you expect to move.

Best regards.

PS. In order to understand what I'm saying, I attached the sketch previously posted
here