Framework

On this page

Analyzing tubular KT joints with overlaping KTK braces and negative gap value

When analyzing tubular joints that have a negative gap value and KTK braces are overlapping (KTT brace does not touch the chord), the computation of SCF’s may be different than according to the rules. Framework has some limitations for this type of joints so we recommend to use one of two workarounds specified below:

  • You can replace the KTT with non-tubular profile, then framework will handle this as a pure K-joint;
  • Alternatively, you can use the “assign SCF” local option to manually assign the SCF value to be used in the calculations;

Back to top

Assign SCF at the member end as Butt-Weld, why is global SCF used?

There are several restrictions on SCF assignment. The PARAMETER SCF rule is usually applied for tubular joint connection. It is valid only at member’s end positions. This option is greyed out if the selected position located in the middle of the member. By contrast, assigning BUTT-WELD or CONE-TRANSITION SCF to a position at member’s end won’t work. If the CONE-TRANSITION alternative is assigned to a transition with no true cone junction, the SCF calculation will fail and the global axial SCF will be used. A message similar to below will be given in the mlg file.

Back to top

Average ring separation when only one ring stiffener is specified

From the Smedly and Fisher paper on ring stiffened tubular joints, it is given that the average ring separation is defined as p = 2* distance of ring to saddle when only one ring stiffener is used. Currently, the Framework user manual is stating that the ring stiffener implementation is according to the Smedly/Fisher paper. However, the above mentioned (when only having one ring) average ring separation is not automatically calculated. If the user wants to specify this, then it has to be done manually.

Back to top

Error message in wind fatigue analysis

Question

What may make Framework issue the following message in a wind fatigue analysis?

Warning from GRES73 called by GWRSTL:Reference outside pointer tableRecord identifier: BELOAD1

What Framework does

For a model containing a flare boom and a part of the deck, if non-zero air drag coefficients Cdy and Cdz are applied on flare boom beams and zero Cdy and Cdz are applied to deck beams, Wajac will only apply wind loads on beams with non-zero Cdy and Cdz. When the result file R1.SIN is imported into Framework, Framework will stop to read in loads when a beam without any wind loads is found, and the above message will be issued. However, the analysis will continue in the absence of other wind load cases and results will be reported. The calculated results are incorrect.

Solution

All FE elements included in a wind fatigue analysis must have non-zero wind loads applied on them. If wind loads on some elements are not desired, very small air drag coefficients, such as Cdy = 0.001 and Cdz = 0.001, can be applied to deck beams and flare boom support beams to make Framework read in wind loads correctly and generate the correct analysis results.

Back to top

How does Framework perform directional combination in a spectral earthquake analysis?

Question

How does Framework perform directional combination in a spectral earthquake analysis?

Solution

The following modal combination methods are available in Framework:

  • CQC – Complete quadratic combination method
  • SRSS – Square root of sum of square
  • ABS – Absolute sum of each modal response
  • NRL – Naval research laboratory method
  • APIC – The method recommended in API RP-2A. In Framework it is the same with CQC method

Then Framework will perform the directional combination based on the selected modal combination method as below.

Modal Combination        Directional Combination
CQC                      SRSS
SRSS                     SRSS
ABS                      ABS
NRL                      SRSS
APIC                     SRSS

Back to top

How to create a set for members or joints in Framework?

Question

How to create a set for members or joints in Framework?

Solution

User can use the command SELECT SET to create a set for members or joints in Framework. The following commands create the set SET1_BMS for members. Set SET1_BMS contains four members.

SELECT SET MEMBERS SET1_BMS ONLY BM1
SELECT SET MEMBERS SET1_BMS INCLUDE BM2
SELECT SET MEMBERS SET1_BMS INCLUDE BM3
SELECT SET MEMBERS SET1_BMS INCLUDE BM4

The following commands create the set SET2_JTS for joints. Set SET2_JTS contains four joints.

SELECT SET JOINTS SET2_JTS ONLY JT1
SELECT SET JOINTS SET2_JTS INCLUDE JT2
SELECT SET JOINTS SET2_JTS INCLUDE JT3
SELECT SET JOINTS SET2_JTS INCLUDE JT4

Here are more common options of selecting members or joints to create a set.

%%                Two joints on the line Tolerance
SELECT SET MEMBERS SET_BMS1 LINE JT1 JT9 1.0
%%                  Three joints on the plane Tolerance
SELECT SET MEMBERS SET_BMS2 PLANE JT1 JT3 JT9 1.0
%%                                    -X   +X    -Y   +Y    -Z   +Z
SELECT SET MEMBERS SET_BMS3 VOLUME -27.5 27.5 -27.5 27.5 -28.0 28.0

%%               Two joints on the line Tolerance
SELECT SET JOINTS SET_JTS1 LINE JT1 JT4 1.0
%%                  Three joints on the plane Tolerance
SELECT SET JOINTS SET_JTS2 PLANE JT1 JT3 JT10 1.0
%%                                   -X   +X    -Y   +Y    -Z   +Z
SELECT SET JOINTS SET_JTS3 VOLUME -27.5 27.5 -27.5 27.5 -28.0 28.0

Notes:

  • If a new set name is given, the new set will be created. If an existing set name is given, the existing set will be updated.
  • A member set and a joint set may have the same name, but their definition is not interconnected.

Back to top

How to define the minimum values of joint parametric Stress Concentration Factors?

How to define the minimum values of joint parametric Stress Concentration Factors?

When joint stress concentration factors are calculated from a user selected rule, user can set up the lowest values for calculated SCFs. Then when the calculated SCFs are smaller than the defined minimum SCFs, the minimum SCFs will be used in fatigue analyses.

Below commands can be used to define minimum values for joint parametric SCFs.

Define minimum SCFs in Fatigue Constants

User can define the minimum SCFs in commands of DEFINE FATIGUE-CONSTANTS as below. The defined values will be applied to welds on the chord side and the brace side.

DEFINE FATIGUE-CONSTANTS AXIAL-MINIMUM-SCF 2.5
DEFINE FATIGUE-CONSTANTS IN-PLANE-MINIMUM-SCF 2.5** **
DEFINE FATIGUE-CONSTANTS OUT-OF-PLANE-MINIMUM 2.5

These commands are valid for SCF rules of Efthymiou, Lotsberg, Kuang, Lloyds, and Wordsworth.

Define minimum SCFs for welds on the chord side and on the brace side separately

User also can define different minimum SCFs for the weld on the chord side and the weld on the brace side using the commands DEFINE PARAMETRIC-SCF. When the separation option is ON, the values defined in the below commands are valid, and the values defined in the command DEFINE FATIGUE-CONSTANTS are ignored.

DEFINE PARAMETRIC-SCF CHORD-BRACE-SEPARATE **ON **
DEFINE PARAMETRIC-SCF CHORD-AXIAL-CROWN 2.5
DEFINE PARAMETRIC-SCF CHORD-AXIAL-SADDLE 2.5
DEFINE PARAMETRIC-SCF CHORD-IPB-CROWN 2.5  
DEFINE PARAMETRIC-SCF CHORD-OPB-SADDLE 2.5
DEFINE PARAMETRIC-SCF BRACE-AXIAL-CROWN 2.5  
DEFINE PARAMETRIC-SCF BRACE-AXIAL-SADDLE 2.5
DEFINE PARAMETRIC-SCF BRACE-IPB-CROWN 2.5
DEFINE PARAMETRIC-SCF BRACE-OPB-SADDLE 2.5

These commands are only available for SCF rules of Efthymoiu, Lotsberg, and Lloyds.

If the separation option is ON, and the rule Kuang or Wordsworth is selected, the defined minimum SCFs in the commands DEFINE PARAMETRIC-SCF are ignored and the minimum SCFs in DEFINE FATIGUE-CONSTANTS are applied.

Back to top

How to enable stochastic fatigue for a submodel with no local wave loads?

Including local loads like wave loads is normally recommended in a submodel analysis. However, if the wave loads are not important for the submodel analysis, the local wave load analysis may be skipped. In that case the file S#.FEM, required for stochastic fatigue, needs to be created for the submodel.

The S#.FEM file has to be present when Sestra is run for the submodel, together with the T#.FEM file(s) - and L#.FEM file(s) for possible local loads. If not, no stochastic fatigue can be computed later.

As the wave load cases for the submodel have to match the wave load cases for the global model, the global S#.FEM file (e.g. S1.FEM) can be copied and renamed to the submodel superelement number (e.g. S4.FEM). Renaming the file is okay because the contents are Sestra commands. (Renaming a T#.FEM model file is not okay.)

Please notice that the contents of the S#.FEM file are described in the Sestra user manual.

Back to top

How to export pictures from Framework, Profast, Stofat, Postresp?

Update November 2017:As Office has been more restrictive to Postscript due to a security vulnerability in EPS please try this workaround:select "Microsoft print to PDF" or a similar PDF printer driver as default printer on your computerinside Postresp set the Plot to Format Windows-printersWhen you Plot the image, this will be saved in a pdf file

Please note that the file will not be properly saved until you close Postresp.

It was possible to import CGM files directly into MS Office in the past. However, a Windows update rendered MS Office uncapable of recognizing this file format.

In order to export pictures from the software it is recommended to export as Postscript format and change the file extension from .ps to .eps.

Then it is possible to import into MS Office.

Back to top

How to generate a PDF file for a plot from Framework, Profast, Stofat, Postresp

User can follow the below steps to create a PDF file for a Framework plot.

  • After a desired plot displayed on the screen, on **Set **tab select **Plot **and then choose Postscript. Click Apply or OK to save the selection.
  • Click Plot, the file with the defined name, such as FrameworkActivity1.PS, is created in the analysis folder. Now exit Framework.
  • In the analysis folder, double click the generated postscript file FrameworkActivity1.PS to open Acrobat, click YES to generate the PDF file.

See also the attached document for details.

Back to top

Influence function methods A and C for SCF computation in stochastic fatigue

Version: All

Description: There may be a big difference in damage results between Efthymiou methods A and C for stochastic fatigue analysis.

Explanation: First of all Efthymiou method C is a conventional SCF approach and not an influence function formulation, while Efthymiou method A and B are influence function formulations where method A includes multiplanar effects and method B does not.

It should be noted that joint load path joint used in connection with stochastic fatigue analysis is not recommended. The reason for this is that the force flow used is based on one wave phase step. The phase used is the first phase angle defined by the command DEFINE CONSTANTS PHASE-ANGLE (default = 0.0). This is a limitation in the load path approach, see status list of Framework case “Efthymiou's parametric SCFs, stochastic fatigue case”, Oct. 2006 and case  “Stochastic fatigue and joint classification”, Dec. 2012.

The difference in results reported for Efthymiou method A and C is a result of the influence function formulation. In the influence formulation the influence braces contributes to the final SCFs applied in the analysis where contributions from the influence braces are added to SCFs of reference brace, see described in Appendix C of the user manual of Framework.  The influence parts of the SCF functions are multiplied with the factors FB/FA and MzB/ MzA, where FB and MzB are the axial force  and out-of-plane moment of the influence brace and FA and MzA for the reference brace.

Assume SCFs are calculated at the two brace ends for every frequency and for all wave directions, a total of 360 load cases. Forces and moments vary for all the load cases. It appears that at some frequencies the axial force and the out-of-plane bending moment of the reference brace are very small compared to forces and moments of the influence braces. Due to this, large SCFs for the influence braces are obtained resulting in large resulting SCFs. Two such load cases are repeated below where SCFa is SCFs of reference brace, SCFsumb is sum SCFs of the influence braces and SCFa+b is the resulting SCFs, i.e  SCFa + SCFsumb.

By imposing upper limits to the force and moment factors, the resulting damage values are affected very much. For an example case using method A it was observed that an upper limit of 10 for the factors gives a damage of 0.288 while with no upper limit a damage of 67.8 is obtained.

It should be noted that in the paper of Efthymiou “Development of SCF formulae and generalized influence functions for use in fatigue analysis” good agreements with experimental results are reported for stress indexes, which is equivalent to the force factors, up to a value of 10. Expected results above this level is not commented, but some test analysis and others too indicates that large force factors (stress indexes) due to small force and moment values in the reference brace may result in unexpected high damage values. This has been observed in deterministic fatigue analysis when stepping through a wave. A step may be such that forces in the reference brace close to zero may occur resulting in large SCFs and stress values which again result in a large stress range for the wave load and unexpected high damage values.

In stochastic fatigue the SCFs are multiplied with the stress transfer functions scaling up the response spectrum and spectral moments on which the damage calculation is based on. In a stochastic analysis the number of calculated stress cycles will change with values of the SCFs. The reason for this is that the calculation is based on the spectral moments which again are affected by SCFs through integration of the stress response spectra. The number of cycles is given by:

Cyc = prob*Timesec/ resTz where
ResTz = 2p*sqrt(m0/m2) = Upcrossing period m0,m2 = spectral moments
Timesec = Year* 3.1536E7 (design life in seconds)
Prob = sea state probability * wave probability

The above parameters can be obtained in a dump file and the number of stress cycles may thus easily be verified. For an example case, the dump file shows that the spectral moments are more than ten times bigger for method A than for method C. The upcrossing period for method A is also larger for method C, giving larger number of stress cycles for method C compared to method A. The reason for this is the m0/m2 relationship for the two methods.

* ---New load step---
* Forces :      fxa       rmya     rmza
*        : 5.71807E-02 0.284585 4.177515E-02
* SCFa   : 11.4196 8.34104 5.67013 5.39434 2.13929 3.98445 2.90565 3.69266
* Influence planar:
* BM28
* Forces :    fxb     rmyb     rmzb  fxb/fxa rmzb/rmza
*        : -0.693008 -2.469662E-02 0.408301 -12.1196 9.77377
* SCFinfb: -141.118 -89.1023 -51.0927 -46.4328 0.0 33.0676 0.0 30.646
* Influence multiplanar:
* BM20
* Force  :    fxb   fxb/fxa
*        : -8.40527 -146.995
* SCFinfb: -406.578 -1.009854E-05 -24.3838 -4.039416E-06 0.0 0.0 0.0 0.0
* BM17
* Force  :    fxb   fxb/fxa
*        : 0.142392 2.49022
* SCFinfb: 10.8958 2.706286E-07 0.653456 1.082514E-07 0.0 0.0 0.0 0.0
* BM19
* Force  :    fxb   fxb/fxa
*        : 10.165 177.77
* SCFinfb: 491.691 1.221257E-05 29.4883 4.885029E-06 0.0 0.0 0.0 0.0
* SCFsumb: -45.1088 -89.1023 -45.3347 -46.4328 0.0 33.0676 0.0 30.646
* SCFa+b : -33.6892 -80.7613 -39.6646 -41.0385 2.13929 37.052 2.90565 34.3386

* ---New load step---
* Forces :       fxa         rmya         rmza
*        : -0.477004 8.682181E-02 3.121898E-03
* SCFa   : 11.4196 8.34104 5.67013 5.39434 2.13929 3.98445 2.90565 3.69266
* Influence planar:
* BM28
* Forces :      fxb        rmyb     rmzb  fxb/fxa  rmzb/rmza
*        : 0.304695 5.58597E-02 0.714325 -0.638768 228.811
* SCFinfb: -7.43766 -4.69617 -2.69285 -2.44726 0.0 774.135 0.0 717.444
* Influence multiplanar:
* BM20
* Force  :    fxb   fxb/fxa
*        : 5.61705 -11.7757
* SCFinfb: -32.5708 -8.089888E-07 -1.95337 -3.235955E-07 0.0 0.0 0.0 0.0
* BM17
* Force  :    fxb   fxb/fxa
*        : -1.6614 3.48298
* SCFinfb: 15.2396 3.785192E-07 0.913967 1.514077E-07 0.0 0.0 0.0 0.0
* BM19
* Force  :    fxb   fxb/fxa
*        : -6.80051 14.2567
* SCFinfb: 39.4324 9.794185E-07 2.36489 3.917674E-07 0.0 0.0 0.0 0.0
* SCFsumb: 14.6636 -4.69616 -1.36737 -2.44726 0.0 774.135 0.0 717.444
* SCFa+b : 26.0832 3.64488 4.30276 2.94708 2.13929 778.12 2.90565 721.136

Back to top

Read jnl file (Postresp/Stofat/Framework)

A JNL command file cannot be read from within Postresp (or Stofat, Framework) when it is located in a folder using blanks (spaces) in the name or path. Workaround: copy the file to another folder, e.g. C:\DNV\workspaces

Back to top

Stress unit is not SI (N/M**2) at Position xx

Version: All from 3.12-00

This message was introduced in Framework V3.12, and is there to warn the users of potential issues in the model when using SCF calculations.

Assume a model for some reason has a few members with Young's modulus 24.0e12 Pa. Framework will by default consider E=2.1e12 Pa. The ratio of E will be applied to factor the stress unit of SN curves, which impacts the damage result. This message is to alert the user that the stress unit has been scaled and that one should double check whether these members indeed have the correct properties.

Back to top

What does the time history fatigue analysis dump file contain?

Version: All

The time history fatigue dump is shortly explained under the command Define Fatigue-Dump in the Framework user manual. In more detail, the file contains the following:

The fatigue dump file for time history fatigue first contains the stress ranges for each beam member in the structure. For a certain member, first it states the member data, which basically explains which fatigue check positions are there in the beam and which SN curve is used. Next, it includes a print of all the stress ranges for the beam, for each hotspot. The largest stress ranges of each hotspot are in the top of the list, and the list then continues down until each stress range, resulting from the rainflow counting, has been listed. Below the table are some short statistics about the stress ranges for each hotspot. The following table is for the next fatigue check position in the beam, etc., until the data for all beams in the structure has been dumped.

Back to top

Why the reported Can Length is not equal to the modelled length in GeniE

Refer to Page 84 in Framework UM, it emphasized that the can/stub segment must be defined with one FEM element. If assign a mesh property to these beams to divide the can segments into two or more FEM elements, this will lead to the incorrect can length is recognized by Framework.  

Use PRINT CHORD-AND-BRACE can check the can length in Framework, the information will be listed as below.

Back to top