Stuck? 
support@mecway.com
support@mecway.com
Mecway
FEA
Axysimmetrical Nonlinear Model
Hello,
I’m taking a look at the new “Direct Route” method proposed by EN13445 Annex B.
I have found that most of the examples and validation files around are Vessels solved with nonlinear axisymmetric solid elements as the method requests an elastic-perfectly plastic material model.
Mecway doesn’t have this feature but it can be easily sorted without too much computational cost.
If one revolves the 2D shell model (tri and quads are acceptable) an RYangle = 1º with a number of subdivisions = 1 and then apply the constrain displacement (-z,0,x) =0 to all the nodes , it provides a very accurate result similar to the ones found in other software’s.
Victor confirms that approach is good and it's what CCX does internally but for some reason it is not extended to Nonlinear.
Time dependent Linearization becomes available also if needed.
The assembly and resolution of the matrices is much faster and more robust imposing displacements (-z,0,x) =0 than frictionless support. (Only drawback is that it doesn’t work for points on the axis of revolution).
I have attached a first comparison between a shell of revolution and the equivalent “Axisymmetric” Nonlinear Model. (Time to solve is 12 min with Pastix)
Hope it helps someone with similar interests so we could share experience/results on EN13445 Annex B.
I’m taking a look at the new “Direct Route” method proposed by EN13445 Annex B.
I have found that most of the examples and validation files around are Vessels solved with nonlinear axisymmetric solid elements as the method requests an elastic-perfectly plastic material model.
Mecway doesn’t have this feature but it can be easily sorted without too much computational cost.
If one revolves the 2D shell model (tri and quads are acceptable) an RYangle = 1º with a number of subdivisions = 1 and then apply the constrain displacement (-z,0,x) =0 to all the nodes , it provides a very accurate result similar to the ones found in other software’s.
Victor confirms that approach is good and it's what CCX does internally but for some reason it is not extended to Nonlinear.
Time dependent Linearization becomes available also if needed.
The assembly and resolution of the matrices is much faster and more robust imposing displacements (-z,0,x) =0 than frictionless support. (Only drawback is that it doesn’t work for points on the axis of revolution).
I have attached a first comparison between a shell of revolution and the equivalent “Axisymmetric” Nonlinear Model. (Time to solve is 12 min with Pastix)
Hope it helps someone with similar interests so we could share experience/results on EN13445 Annex B.
Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!
Comments
I am very interested in performing all the validation examples of the new “Direct Route” method proposed by EN13445 Annex B.
I am very new to this software, but I am available to share the load if you like to drive the activity and guide me.
That seems very ambitious!!.
My goal some months ago was to introduce myself to the method and validate some of the examples, but unfortunately, I found some drawbacks that made me stop.
First one , Linear-elastic ideal-plastic constitutive laws are used for the EN inelastic analysis which is not available in Mecway yet. That means there wouldn’t be an easy comparison. I tried to advance on the examples substituting the perfect plasticity with a bilinear isotropic hardening model with very low Tangent modulus.
I found some strange behaviours I couldn’t explain or get response from the Ccx forum. (My equivalent plastic strain was diverging in areas where there was a sign reversal in the stresses during the cyclic loading).
Maybe it’s time for a second chance. ¿Are you following any book, manual or directly the code?. I read the code (very dark) and then followed the book “Pressure Vessel Design - The Direct Route (Zeman-Ed. Elsevier) “
Apart from that, the Axisymmetric approximation using the revolved solid elements perform very well in linear elastic at least. I have obtained very good agreement with EN 13445:2014+A8:2019 (2020-07) formulation.
I have managed to solve two of the initial difficulties before facing again the application of the Method proposed by the EN code. ( Perfect-Plasticity and Yield Criterion)
1-Perfect Plasticity can be set up in MECWAY by means of custom cards. The method requires a good control and avoidance of high peak stresses. If not, the convergence with perfect plasticity or any other model becomes difficult.
2-EN13445 AnnexB Gross plastic deformation design check, B.8.2 requires a linear-elastic ideal-plastic law with Tresca's yield condition (maximum shear stress condition) and associated flow rule is required.
Plastic flow algorithms in Calculix are isochoric and the classical von Mises Huber yield condition applies but there is a note on the EN13445 code that could be used.
“If no subroutine for Tresca's yield condition is available or if it shows too bad convergence, Mises' yield condition can be used instead. Since the maximum ratio of the Mises equivalent stress to the Tresca equivalent stress for the same load is 2/V3 , a multiplication of the design resistance with V3/2 will always lead to conservative results”.
Strenght Parameter in the Elasto Plastic material properties must be corrected RM = 255Mpa / 1,25 *sqrt(3)/2.
I’m working on the “Example 1.2 Thin unwelded flat end” found in the “The design-by-analysis manual” if you are interested [1]. I,m using C3D8R elements as suggested in the Calculix Forum.
The manual solved it with the compensation method but it can be used as reference for its simplicity. (Example number 2 could also be interesting at least for me).
If I found a way to apply the Compensation Method with Calculix it would be TOP. (Any hint is appreciated).
Still looking how to sep up each of the Analisys and details for further comparision.
[1] https://op.europa.eu/en/publication-detail/-/publication/46acbdec-5efe-11e8-ab9c-01aa75ed71a1
I am terribly sorry for my shameful delay in replying.
I'm finally able to start from scratch to use and test Mecway.
Is there a way to contact you directly when I'm ready to help?