converging-divergent variable duct pressure

[delphy90]

hello to all are struggling with structural analysis on a convergent-divergent conduit.
Can someone suggest me how to apply a variable pressure inside it, knowing only the values at the entrance, in the throat section and at the exit? Do you recommend using lbc fileds?
thanks to alliiii

 

[Onda]

If you know the pressure field to apply, you can create a field in the variable corresponding to the axis of the duct.
For example, if the conduit has x axis, your field can be: ♪
where a and b are constant in this way you can create a linear pressure variation.
ovviamnete, for a conduit you will have to create two field, one for the convergent area, and one for the divergent area, so with a variation of the quota in x, first the pressure grows and then decreases.
more precisely, for a linear variation normally the field takes shape
a + b* ('x - x0)

 

[delphy90]

if the conduit has x axis, your field can be: ♪
where a and b are constant in this way you can create a linear pressure variation.

the problem is to identify the two constants. I can't do it because I don't know the parts of the points. I know only the punctual values, that is the value that the pressure takes at that point of space.

 

[Onda]

then you do not have to solve the problem with nastran but with a cfd.
you must first find the pressure field, then apply it to do a structural analysis.
Otherwise, if you do not want to resort to a cfd you can find pressure diagrams in the ducts, calculate the pressure trend and then do the analysis.

 

[delphy90]

Otherwise, if you do not want to resort to a cfd you can find pressure diagrams in the ducts, calculate the pressure trend and then do the analysis.

how to calculate the pressure flow through the diagrams? set a tabular field in an experimental and approximate manner?

 

[Onda]

I still don't understand.
If you have to do a structural analysis, you have to have pressure fields, otherwise you first do a cfd analysis and then do that structural.

To solve the pressure in a nozzle, it is done with equations, I would say bernoulli. with this you can calculate the pressure field at each point and get a formula, which then through a field, apply to your elements.
I wouldn't call it an experimental and approximate system.
experimental means you do a nozzle and try it in the wind tunnel by putting probes in various points to get the pressure.
Approximate, that means you put the loads randomly.
If you use the bernoulal equation, you can calculate in each point the pressure and use the formula to do a field and apply the load on your elements.

 

[delphy90]

I would like to ask the Commissioner whether he is a member of the Commission.
what you've described can I do in Patran himself?

 

[Onda]

I don't think it's the best tool.
I would do in excel the calculation of pressures for discreet sections.
basically, do a section of your nozzle every tot mm.
in excel do a column with the areas of the section, and solve the speed for the continuity equation.
find the speed with bernoulli find the pressure.
Then create an interpolation curve and this one puts it in patran in a field. I wouldn't infiltrate a cfd, unless I had so much desire and time.
Of course bernoulls work if the fluid is incomprehensible. Otherwise you have to use other equations, it's not exactly my field that of compressible fluids.

 

[delphy90]

It could be the best way. I will try to follow your intuition.. thank you very much!! !
You gave me valuable advice!

 

[Onda]

Please.

should be simple enough.