08-07-2016 03:03 PM
Dear all,
I would like to performed an explicit analysis in femap. FIrst of all I have a problem to define a material coefficient for my material (the coefficient required in creep card of oak ridge national laboratory).
The calculation code required an Omega method for explicit creep analyis: how can define this material law in femap?
Regards
Stefano
11-23-2016 08:05 AM
Hm, as far as I know, creep is a very long-term effect, and as such, is more suited to quasistatic transient analysis, which is typically an implicit solution with very large time steps.
Is the material concrete?
11-23-2016 05:52 PM
Dear Kava,
no is steel material for pressure vessel design
11-28-2016 10:01 AM
By Oak Ridge, do you mean some kind of LS-Dyna input?
My gut feeling is that the boiler codes should all specify the creep properties of common steels, or at least reference where that information is to be obtained. Usually, this is input as a creep rate (strain per hour/day/whatever) vs stress AND temperature, so a 3D graph basically.
I've never done this in FEMAP, but I have done creep analyses for pressure vessels before in Strand7.
In FEMAP it is defined in the Material, under the Creep tab.
If you can't find the relevant data in your standard, try googling creep curve for the material you are using.
12-08-2016 04:57 PM - edited 12-08-2016 06:40 PM
Hi Stefano,
I haven't tried this myself, but am also interested in exploring this area.
I suggest you take a look at the reference material accessible in the FEMAP menu bar under Help - NX Nastran.
For explicit analysis look at the "Advanced Nonlinear Theory and Modeling Guide" this of course requires access to the NX Nastran Advanced Solvers.
But it looks like you could use the Basic Nonlinear Solver (Sol106) to perform an implicit creep analysis. See "Basic Nonlinear Analysis User's Guide" section 3.5 and "Handbook of Nonlinear Analysis" in section 6.4.7 and 6.4.8 there is an example for pressure vessel head and nozzle from 304 stainless steel.
According to these manuals, there are two creep laws available in Sol106. For creep law class 1, I believe the following reference is applicable http://www.osti.gov/scitech/servlets/purl/6682954 (NEF 9-5T "Guidelines and Procedures for Design of Class 1 Elevated Temperature Nuclear System Components"). I haven't been able to find a definitive source for the ORNL coefficients in one reference, but various test data and examples have been published in reports available from http://www.osti.gov/scitech/.
Creep law class 1 seems specific to Austenitic Stainless Steels (particularly 304 and 316), and possibly Cr-Mo steels (I haven't checked this out in in detail).
Does anyone have more recent references for this data ?
If the problem you are looking at can be solved in simple constant temperature steps, I believe you could:
You'd need to be careful to check that this creep law (type=300), accurately fits the data for the material you are trying to represent (e.g. by plotting both in excel and comparing curves). This creep law is the Bailey-Norton law (classic power creep law).
Theoretically I think it should be possible to directly utilise the Omega data, but it looks like this would require Nastran DMAP programming to implement a user subroutine inside the Nastran Solution Algorithm. Is this correct and has anyone attempted this ?
Regards
Fergus