Applies To Product(s): AutoPIPE, Version(s): 2004, XM, & V8i Environment: N/A Area: Modeling Subarea: Original Author: Bentley Technical Support Group Attention: Please see the following AutoPIPE help section: Help Contents Contents Tab Modeling Approaches Modeling Approaches This help has been provided in order to give users ideas for modeling typical piping arrangements. The steps shown in each example should not be taken as the only method available to create models. In addition, the intent of the examples is to present ways to create adequate models of specific piping components for analytical purposes. Anchor Bends Cuts: Cold Spring Flexible Joints Frames Hangers Nozzles Pipes Reducers Rotating Equipment Supports Tees Valves Vessels Comments, Questions, and Answers: Item #1 For flexible joints, how to apply define coefficient of thermal expansion, and where it is defined? Answer: The flexible joint will have an effect on the thermal movement of the pipe based on the stiffness specified on the flexible joint dialog screen. In addition, translation of thermal movement can be accomplished by adding a axial restraint across the flexible joint. Another words, a flexible joint itself does not undergo any thermal expansion/contraction i.e. no thermal properties are applied to the joint itself. As seen in the online help: Flexible joints are modeled as a one-point lumped spring expanded into a two-point spring element. Item #2: Where can I input the material of the flexible joint? Answer: AutoPIPE lets the user define the stiffness terms of the flexible joint directly so no material property definition is needed. Therefore the following 4 statements are true for Flexible Joints: A. No thermal expansion is applied to the joint B. No material property is applied to the joint. C. Code stress color plot does not apply to flex joints. D. Hoop stress or other code calculations does not seem to be calculated for flexible joint. Item #3: Wh en modeling the Tied-belows example using Tie/Link supports. The tie/links allow the pipe to move father apart than the length of the tie/link. Answer: Correct a tie/link support does not maintain arc swing length. However, they do maintain the distance between tie rod ends where the main control function is to maintain the flanges parallel. No program including Caesar will arc tie rods and even if it does would be insignificant change in the expansion joint behavior. Item #4: Can you send me an example model of a bellows than what is available in the online help: Answer: yes, see the following: http://communities.bentley.com/products/pipe_stress_analysis/m/pipe_stress_analysis_gallery/260260.aspx Item #5: How do I model Flex / Braided hose: Answer: Flex hoses are typically so flexible that it is like a break in the pipe with no transfer of force or moment therefore a simple axial bellows with low stiffnesses in all degrees of freedom. Unless the the manufacturer can provide some stiffness values. Suggest using 1 or multiple flexible joints back to back with tie rods as required. If no stiffness is given, I would recommend using a zero stiffness for the flexible joint, but due to possible instability, you would need to provide a small non-zero value, e.g. 1 lb/in. I would also recommend that you set the pressure area to zero. You may be able to check for bending radius if you calculate an equivalent angle. Frame elements may be used, but flexible joints are easier in this case. For frame you would need to enter non-standard section properties instead of stiffness. Note: remember that AutoPIPE cannot handle large deformation and forces/moments are based on original un-deformed geometry. Item #6: When modeling a flexible hose, what is the most realistic way to model the curved flex hose. Using 1 flex joint, 2 flex joints or many flex joints? Answer: Keep in mind what a flexible hose is doing, as mentioned above "it is like a break in the pipe with no transfer of force or moments". Therefore you can model it using a single flexible joint or using multiple flexible joints where both approaches would use very low Axial, Bending, and Shear stiffness values. Assuming the stiffness values used are very low, then both modeling approaches would be valid. The only major difference between them would be the CG location of the hose. Depending on the hose size, it may need to be considered. Thus giving a slight edge to using multiple flexible joints when modeling a hose (assuming you know the exact path of the flexible hose during operating load cases). Note: 1. When inserting a flexible joint be aware of the stiffness directions and how that affects the connected pipe. Example, using a single flexible joint, axial stiffness is acting along the line of the connected points. Using low values as mentioned above, should not affect the results very much. However, if for what ever reasons axial stiffnesses are higher, it could affect how the connect pipe moves during load case combinations. 2. If modeling multiple flexible joints back to back. Consider adding a short bend pipe between the joints to reduce the occurrence of warnings (i.e. W726-8: Kink in straight run ...). Item #7: On the flexible joint dialog screen, "Pressure thrust area" From the online help: This is the effective cross section area, usually based on the mean diameter of the convolutions of the expansion joint. It is multiplied by the internal pressure to obtain the axial thrust due to internal pressure. This thrust is used if a rigorous pressure extension analysis is requested." If I understand this text correctly, this thrust means a force that has to be applied to the pipe at both ends of the expansion joint due to the pressure in the pipe. The direction of this force is axial and away from expansion joint. Please confirm my understanding. Also, how do perform a "rigorous pressure extension analysis" in AutoPIPE Answer: Correct - a force is applied in each direction. Please see the following AutoPIPE help section: Help Contents Search Tab enter "rigorous pressure" (include the quotes), press List Topics button, double click on the " Include Axial, Pcase in Sustained" topic from the list provided to see more information on this option. Item #8: How to model a Penetration Seal Bellows Joint in Autopipe (see image below). This Seal Bellow Joint is welded to a pipe at one end, and to a seal plate wall penetration at the other end. The pressure inside the Penetration Seal Bellows is P(PSB)= 0.1 MPa, and the pressure inside the pipe is P(Pipe)= 3 MPa. The Penetration Seal Bellows Joint must be able to accommodate both the axial (DX), vertical (DY), horizontal (DZ) displacements, and the angular/torsional (RX) rotations. Answer: Using the same techniques used from modeling Jacketed piping: 1. insert a set of node points on the existing pipeline exactly where the Flexible joint is to be modeled, (ex A175 and A180) 2. Select node point that matches wall location.(ex. A175), insert a new Segment (ex. C), offset from A175, a small distance (ex. 0.01 ft). with a new PIPEid. 3. Insert a new PIpeID that matches the size of the expansion joint, (ex "14std"). 4. insert Flexible Joint (ex C00 to C01) that matches the same distance as as previous node point set (ex. A175 to A180). Finish entering the correct settings as needed for the flexible joint dialog and press OK button. 5. Insert a Rigid beam between end point on flexible joint to main pipeline. (ex. C01 to A180). 6. Select First node point on flexible joint (ex. C00), insert rigid anchor. 7. Done, should look something like this: Item #8 How to model a Hyspan Barco Ball Joint? Answer: Please see the following AutoPIPE help section: Help Contents Contents Tab Modeling Approaches Modeling Approaches Flexible Joints Ball and Socket Joint example in AutoPIPE's online help. Actual ball and socket joints are limited in their range of angular rotation. AutoPIPE will not limit this range. Therefore, they should be placed in the piping system so that these limits are not exceeded. Generally, the vendor provides a set of break-away torques for the ball joint assembly. For this, I would recommend modeling the ball joint with a breakaway stiffness that will be constant throughout the travel. This will be conservative. There is a limit load reached before the joint goes to 'zero' stiffness. You can check the Forces and Moments report at the joint to see if this load is reached. If the load has not been reached, increase the corresponding stiffness and recheck. If load has been exceeded, decrease stiffness accordingly and recheck until breakaway forces have been approximated. This issue has been logged as a requested program enhancement with tracking number CAE-TR-4779. See items #9 and #10 for comparative approach. Item #9: how to best model the bending stiffness and torsional stiffness of the ball joints. The manufacturer has given the following values for the ball joints; Axial stiffness = Rigid Transverse shear stiffness = Rigid Torsional resisting moment = 4610 ft -lbs Bending resisting moment = 2305 ft - lbs The problem is, that Autopipe accepts a "stiffness" for torsional and bending resistance but the ball joint friction is more a constant "friction torque" during movement and not proportional to the angular displacement. The only solution we can think of is to estimate the amount of angular displacement of the ball joint and scale the stiffness value accordingly so that the friction torque specified above is not exceeded. Can you provide any suggestion on the best way to model this type of joint? Answer: The approach suggested above is reasonable. Another possible solution is to calculate the moment in the joint and if the moment exceeds the limit friction moment. If exceeded, apply a constant moment (using force/moment) on one end of the joint to counter the friction and set rotational stiffness to zero. If the moment is not exceeded, set the stiffness to rigid. (So set to rigid first to estimate the moment and if exceeded, set back to zero with constant moment). 1. Set stiffness to rigid. 2. Analyze to calculate moment in the joint (for all basic load cases, not combinations) 3. If moment does not exceed friction moment, you are done, rigid is valid for that load case 4. If moment exceeds friction. Set stiffness to zero (unless one is provided) and apply a constant moment equal to friction moment on one end of the joint. The moment is applied in the load case evaluated. Make sure the moment sign is correct. The problem with this approach is that you cannot set stiffness to zero for one load case and rigid for different load case. See items #8 and #10 for comparative approach. Item #10: My specific Ball Joints have a particular torque which allows them to move. For example I am modeling a 14" ball joint with a friction torque of 9,000 FT-LBS. Once this moment is developed in the system it remains constant and the ball joint rotates. Answer: First, start by modeling a ball joint as specified by the online help by using a Flexible Joint. Second, currently there is no easy way to model this. One workaround would be to simulate a single load case: Assume joint is rigid. Run and see what torque you have. If torque is less than friction value, you got your answer. If torque is greater than friction value, set stiffness to zero or use actual stiffness and apply an external moment equal to friction torque in the particular load case. See items #8 and #9 for comparative approach. See Also Bentley AutoPIPE External Links Bentley Technical Support KnowledgeBase Bentley LEARN Server Comments or Corrections? Bentley's Technical Support Group requests that you please submit any comments you have on this Wiki article to the "Comments" area below. THANK YOU!
↧