Hi Ms/MR,
When i calculate a pipe system connected with tank,i used the nozzle flexibility-api650 check,but to my surprise,the output of the displacement for nozzle check has two question:
1,The radial growth displacement of the shell is only considered the thermal dispalcement,it doesn't consider the displament for tank deformation itslef.
In API650-P presents W=[(9.8*10-6*GHR2)/(Et)]*[1-e-βLcos(βL)-L/H]+αR△T
2,the unrestrained shell rotation resulting from product head is zero,it is impossiable,in API650-P it presents θ=[(9.8*10-6*GHR2)/(Et)]*{1/H-βe-βl[ cos(βL)+sin(βL)]}.
Then i calculated these two parameter by means of API-650 by hand.The outcome is different.
Now i am confused with this nozzle check tools,why is it fail to check nozzle flexiblity? Are there anyone encounter same problem with me?
I look forward anyone's reply,tks a lot.
Best wishes.

The API-650 nozzle check tool lumps thermal displacement and hydraulic displacement and rotation into a singular calculation. The results appear when you instruct CAESAR to start the run.

If you would like hydraulic displacement separate from thermal displacement, then you would specify a thermal differential of 0.

It's noteworthy that the new verbiage in the B31.3 piping code (if applicable to your case) regarding piping lifting off supports could equally be construed to the hydrostatic effects of having a filled tank, but I would be inclined to use the "normal liquid level" as that value, and not necessarily the full tank level.

If you are expecting CAESAR to automatically apply these displacements at the node in question, you have to provide CAESAR with a displacement (you can set the numbers to 0) at the node to overwrite with the calculated values. Alternatively, you can manually set the displacement at the tank wall by placing an anchor and specifying a CNODE to node that's unused elsewhere in the model, and specifying the displacement at that otherwise unused node.

See attached screenshots.

[img]http://image.baidu.com/search/detail?ct=503316480&z=undefined&tn=baiduimagedetail&ipn=d&word=URL&step_word=&ie=utf-8&in=&cl=2&lm=-1&st=undefined&cs=2293435572,2627456&os=1410355114,3589161492&simid=3204371943,3715087229&pn=2&rn=1&di=25267382580&ln=886&fr=&fmq=1458782743199_R&fm=&ic=undefined&s=undefined&se=&sme=&tab=0&width=&height=&face=undefined&is=&istype=0&ist=&jit=&bdtype=0&gsm=0&objurl=http%3A%2F%2Fwww.qqtn.com%2Fup%2F2015-2%2F201502011119577084883.png&ctd=1458782745816^3_1239X715%1[/img]

Hi Michael,
Tks and grateful for your quick reply,i try the way in your instruction in CAESAR II.The shell rotation resulting from product head θ is exisited in Caesar II output list,but the displacement by hydraulic is still not exiisted in CAESAR II output list.I want to show the picture for these information in above 'Reply',it seems failed.
For above reply which u showed me picture,can u check the Caesar II output list 'Displacements' and show me in the picture.
By the way how do u insert the picture as above u showed me,i try many times but i can't get it.

Hi Michael,
Greeting to u,right now,I am making a tank nozzle check by means of API 650-P in 'office excel',i need calculate 'Axial stiffness KR','longitudinal sitffness KL','Circumferential siffness Kc',they are figured out by parameter'KR/E(2a),KL/E(2a)3,Kc/E(2a)3',then i need to check Figure P.2a which ralated to parameter 'a/R' and 'R/t'.The parameter I calculate a/R=0.008 and R/t=1180,the curve a/R=0.008 is between curve a/R=0.02 and a/R=0.005,but I am hard to find where is the accurate loacation for curve a/R=0.008,it is too indistinct to choose this curve,thus I think when i choose the value only by my visual,it will produce a big error for my calculation.
The another way, i input the basic tank and nozzle datas in Caesar II,and it automaticlly calculate, i find the output for 'Axial stiffness KR','longitudinal sitffness KL','Circumferential siffness Kc' are different from the values which i calculate by excel tool.
I conclude the method i choose 'KR/E(2a),KL/E(2a)3,Kc/E(2a)3'is different from Caesar II default method.
So that,what the method of Caesar II to choose these parameter?it is also relyed on the picture P.2a-l,P.4a and P.4b in API 650-P? I guess these pictures will be converted into equal equations in Caesar II software,thus it can avoid big error by visual deviation.
I am looking forward to your reply,tks a lot.

One note on that screen shot...
If you wish to see those displacements included in the analysis you must declare a displacement set (e.g., D1) for them and then include that displacement set in the appropriate load case(s). Use the last cell in the window to do so (Set Displacement Vector).

Song,

To attach images directly, look for the "Switch to Full Reply Screen" button below. Simply attach the image files and they will render at the end of your message. I am incapable of seeing the image you linked to, however in MY image, you can see "DEFLECTION." This value includes thermal and hydraulic deflection. However, if you instruct CAESAR to have a "temperature change" of 0, then all you will have left is hydraulic.

Generally in pipe stress modeling when attaching to equipment, you have three options:

1 The equipment is rigid and the pipe is flexible - equipment moves the pipe.
2 The pipe is rigid and the equipment is flexible - pipe moves the equipment.
3 Both pipe and equipment are of similar rigidity/flexibility.

You should strive for option 1 whenever possible, as it does not require the analysis of nozzle stiffness. Reported stresses and nozzle reactions should be conservative.

My opinion is you only need the nozzle stiffness values if you expect the pipe to offer significant resistance or impart significant load into the tank. I would prefer to limit this kind of analysis to existing tanks and existing pipe with poor flexibility characteristics.

Dave,

We ran into a couple challenges on some of our tank models, and the easiest way to be consistent across the board was to manually specify displacements. The short list of complications includes:

• Modeling nozzle displacements in combination with tank settlements. We could not pinpoint why stresses seemed to have been coming out correctly, but reported displacements seemed to be an average between the two.
• Using the Nozzle Flexibility input option and manually specifying the displacements sometimes interfered with each other, so we would ask CAESAR to calculate these values, and then turn the Nozzle Flex off on subsequent runs.
• Nozzle Flexibility will not work if your tank nozzle projects into the tank, although we did not attempt to connect this via CNODE. However, I opted this out of the analysis due to being existing and operating on the assumption that the tank moves the pipe - if there's a problem already, we didn't create it, and we weren't asked to look at existing problems unless they caused failure in items within scope.

Hi Michael,
Tks for your response soon,I will consider your good advice as you said above.
Tks a lot.