line with expansion joint: force on flange and on pump inner surface

i have a straight line with an expansion joint (without tie-rods) connected to a pump flange.

- line pressure is P
- expansion joint "inner hole" surface is S.

Let assume caesar gives force on pump flange node equal to 1000N.

Question: do the "1000N" contain both force on pump flange both the force on inner pump wall or not?

i mean: force on inner pump wall is P*S

is force on pump flange 1000N or 1000-P*S?

In many instances, pressure causes no structural response in CAESAR II analyses - exceptions are with long pipelines and FRP pipe where axial stiffness is low enough to develop significant system deformation. You can consider this structural response due to pressure in other systems by turning on the (axial) "Bourdon effect". On the individual component level, untied expansion joints can also develop axial distortion due to pressure. In the typical CAESAR II model, the force of the pressure times the entire, effective bellows area is applied on either end of the joint rather than on the internal surfaces projected up- and down-stream of the joint. I believe there are several Forum posts on this application.
I suggest you run a load case with pressure alone. The results will show the CAESAR II nozzle loads. My guess is there will be a compressive load on your nozzle if you entered an effective ID (and pressure) for that XJ. But, as you say, this load includes the calculated force carried by the back of the pump housing. To get a more accurate estimate of the nozzle load, I suggest you break the XJ's axial pressure load into two groups: 1) the load based on the ID of the pipe, located on the back wall of the pump housing (or on the impeller), there is a similar load upstream of the XJ, probably on a bend, and 2) the differential of the effective XJ area and the inside area of the pipe applied on either end of the XJ. If you call these loads "F1", the pressure-alone case would now be "P1+F1". If this works for you, then a good load case for your operating loads on the pump would be something like W+P1+T1+F1.
That force due to pressure may either increase or decrease the total load on your nozzle.

really interesting response but still not clear to me.
i needed to confirm to vendor that pump flange load is below vendor max load, so the question was born "what is the real flange load if there is an EJ?"

(note: yes, my line is a GRP one but doesn't matter, is a general question valid also for metal pipes)

if i well understood, actually, using classic load case W+P1+T1, the caesar load on a pump flange node is the sum of two loads:

1)line fluid pressure load based on the ID of the pipe (that really should act on the back of the pump housing and not on pump flange)

2)EJ thrust based on the differential of the effective XJ area and the inside area of the pipe (that really should act on flange)

The flange should "see" only the load 2) but caesar result give a total load 1)+2). is right?

[moreover the "effective ID" data in EJ caesar panel is the force 1) plus a bit more due to part of EJ inner surfaces.]

so if i look at caesar result, the load on flange node is different from the real one due to presence of load 1)

Using the classic operating loads "W+P1+T1" the effective load on flange is "obfuscated" by presence of load 1) that will non act on flange, will not give load on flange but will be carried by the back of the pump housing.

actually i used operating load "W+P1+T1" but in this way i really don't know the real flange load and i can't check the vendor max loads.

So to calcuate the REAL flange load have i to take the flange caesar load and subtract by hand the load 1) that is P1*S where S=area of ID of the pipe, is it right?

---------

OTHERWISE:
i had another idea:
on pump flange node i can insert a Cnode anchor and insert on it a force equal to -P1*S (having opposite direction to direction from EJ to flange)? in this way i subctract the force that act on the back of the pump housing. So caesar flange node give only the real flange load, without the force acting on act on the back of the pump housing.
is it possible to work in this way?

---------
About the additional load case "W+P1+T1+F1"
where:

P1=the load based on the ID of the pipe, located on the back wall of the pump housing (or on the impeller), there is a similar load upstream of the XJ, probably on a bend.

F1=the differential of the effective XJ area and the inside area of the pipe applied on either end of the XJ

but... adding this "F1" i have to delete the "effective ID" data by EJ caesar panel? otherwise i will have an EJ double effect, one by the "effective ID" data and another by the F1 force. Is right?


thank you for you patience, this argument is driving me crazy

Here's a current rule in CAESAR II: If Bourdon Effect is not active in your model and if there is no "Effective Diameter" specified for expansion joints, there will be no structural response (i.e., no load & no movement) associated with a pressure load.
Pressure causes little response in terms of system movement in most metallic piping systems. For many years, the Codes were quiet on the formulation of the stress due to sustained loads and engineers typically threw in a "PD/4t" for the pressure stress component. Here's where your GRP (FRP) designation comes into play.
GRP/FRP pipe is quite flexible in the axial direction and pressure, alone, may cause system distortion. So, if GRP/FRP material is specified in your CAESAR II model, this "Bourdon effect" is automatically included in the analysis.
BUT, our Bourdon term has two components - 1) the axial extension due to pressure and 2) the "Poisson shrinkage" in the axial direction due to hoop pressure. Note that I am calling these strains rather than loads. That's how they are applied in CAESAR II - they look more like a thermal expansion rather than end loads on each element.
You cannot unscramble these eggs. You cannot just use the Effective Diameter to adjust the calculated flange load due to pressure.
What I would do: 1) run an analysis without pressure to get the flange load excluding the pressure term, and 2) model the pressure distribution between the XJ and flange by hand to get the pressure term, and 3) add 1) & 2) to get the total flange load. To model the pressure load distribution by hand I would specify the P*A(in) load on the back of the elbow beyond the XJ from the nozzle (the balancing force is pushing the impellers/back of the pump) and the P*[A(XJ)-A(in)] on either end of the XJ, pointing away from the XJ; do not specify an XJ effective diameter. These loads would be defined as a force set (e.g., F1). Run this force set alone to see how the pressure loads are distributed through your assembly.

"1) run an analysis without pressure to get the flange load excluding the pressure term":

so i obtain flange load caused by thermal cases only.



"2) model the pressure distribution between the XJ and flange by hand " and then "i would specify the P*A(in).........":

to model pressure distribution i insert the 3 forces (1 on elbow and 2 on the EJ sides) but i DON'T set any line pressure inside pipe AND no "Effective Diameter" specified for expansion joint. Is it right?


otherwise, i will have again the pressure contribution (coming from EJ) acting on flange that should really act on pump wall. is it right?

Yes.

The fact that the thrust will act on the pump wall and not the pump flange do not change the problem. Pump will be subject of thrust load that will lead to the same result on the pump alignment, no matter what. there are pumps that are capable to withstand the thrust, other not. Simple you need to get confirmation from pump vendor. Otherwise anything become a math exercise and a way to by-pass the problem.