The latest release of OrcaFlex 9.4 is OrcaFlex 9.4g.
The new features introduced in OrcaFlex 9.4 are described below. For full details please see the OrcaFlex help file: OrcaFlex.chm (3.1 MB).
Extract the contents of the downloaded zip file to a single folder and then run the extracted file called patch.exe.
A more in depth discussion of the major new features introduced in 9.4 is available on our blog.
New in version 9.4g
- Range graphs for incomplete periods (e.g. partially complete simulations) were not showing results for spike logged variables such as tension.
- The outputting to Excel for the post-processing spreadsheet has been greatly speeded up.
These bugs are fixed in version 9.4g.
New in version 9.4f
- Wake Interference was being mis-calculated when a User Specified wake model was used. This problem did not affect the Huse or Blevins wake models.
This bug is fixed in version 9.4f.
New in version 9.4e
- Copying objects to the clipboard in earlier versions of 9.4 could sometimes result in corrupted models. Typically this bug would manifest as objects changing names, apparently at random, e.g. names becoming blank or names like '1', '2', '3' etc.
- In-plane and out-of-plane shear force results were reversed in previous versions. That is the in-plane values were reported instead of out-of-plane values and vice versa.
- The OrcaFlex spreadsheet Get Data instruction was not working for variable data sources.
- Batch processing for OrcaFlex spreadsheets was failing if the simulation files referred to by the spreadsheet did not exist when the spreadsheet was added to the batch list.
These bugs are all fixed in version 9.4e.
New in version 9.4d
Previous versions of the program would sometimes fail to perform rainflow fatigue analyses, reporting Out of memory errors. These errors would typically occur for analyses with large numbers of nodes, thetas, components, for long duration simulation files or on machines with a large number of processors.
The rainflow fatigue analysis implementation has been reorganised to make more reasonable use of system memory. We believe that the analyses will no longer fail due to lack of memory as a result of this change. A side effect of this is that some analyses may be slightly slower than in previous versions, but we feel that it is best to trade a small amount of performance for increased robustness.
New in version 9.4c
The batch script and text data file generation features now reflect the Excel formatting of value cells. So, if a cell is formatted to have, say, 1 decimal place, the corresponding value in the generated script or text data file will also have 1 decimal place.
- The Total End Force magnitude result was calculated as the sum of the protected line end force magnitude and the stiffener end force magnitude. Instead the vector sum should be performed and then the magnitude of this sum returned.
- End Lx, Ly and Lz-Force and Moment results were reported incorrectly for stiffener line ends. The magnitude was correct but the results were reported with respect to the wrong frame of reference. Similarly, Total End Lx, Ly and Lz-Force and Moment results were incorrect because they are based on the erroneous stiffener end load results.
- End Ex, Ey and Ez-Force and Moment results could be reported incorrectly for stiffener line ends, at End B, when torsion is not included. The error occurred only when there was some twist between the end axes orientation and the end node orientation. The magnitude was correct but the results were reported with respect to the wrong frame of reference.
- Mean up-crossing declustering for extreme value statistics was operating incorrectly for time histories with consecutive values that alternated above and below the mean. This would typically only occur when the log interval was exceedingly large.
- Models with line nodes exactly on edges or corners of solids resulted in unstable simulations.
- Simulations for models with wire frame drawings containing a very large number of vertices could, under some circumstances, take significantly longer than in previous versions of the program.
- Setting Hs to zero for the Torsethaugen spectrum resulted in a floating point divide by zero error.
- The batch script and text data file generation features of the OrcaFlex spreadsheet would, with certain input, result in errors of the form: Could not convert variant of type (OleStr) into type (Double). This would typically occur for cells containing numbers formatted as text, or Excel errors such as #VALUE!, #DIV/0! etc.
These bugs are all fixed in version 9.4c.
New in version 9.4b
Some modern Intel processors can run more than one thread per core. For example, many of the Intel Core i7 processors support two threads per core. The terminology logical processors is often used - to give a concrete example, a quad core processor with two threads per core has 8 logical processors, and so on.
Previous versions of OrcaFlex did not, by default, take advantage of this capability because the thread count defaulted to be the number of physical cores. Starting with version 9.4b, the program defaults to using the number of logical processors and so makes full use of available processor resources.
- The install program for 9.4a failed to install the OrcaFlex spreadsheet correctly.
- In version 9.4a, the OrcaFlex spreadsheet did not function if the language selected in the system's regional settings did not match the Excel program language.
These bugs are all fixed in version 9.4b.
New in version 9.4a
- Line contents can now be specified as free-flooding, in which case the line is filled with sea water, up to the instantaneous water surface.
- A slug flow option for line contents modelling has been added. This allows for spatial and temporal variation of contents. The contents flow velocity can also vary with time.
- Axial contents inertia can now, optionally, be excluded from analysis.
- Line contents flow loads (centrifugal and coriolis forces) are now applied in statics. Previously they were ramped during build-up. Applying the loads in statics reduces transients.
- A contents density results variable has been added.
- Line clashing can now occur between segments from a single line. In previous versions clashing only occurred between segments from different lines.
- The Line with Floats Wizard is now available when the base line type has drag which varies with Reynold's number. In order to allow this the Wizard now sets the drag diameters for the equivalent line type to be the drag diameters of the base line type and sets the drag coefficients accordingly.
- The Maximum Tension data item for Line Types has been renamed Allowable Tension.
- Minimum Bend Radius variable data items can now specify that bend radius is related to either wall tension or effective tension. Previously the relationship was always with wall tension.
- A Strouhal Frequency results variable has been added.
- Time domain VIV models now report VIV results for sections of line that are protected by bend stiffeners.
- Generation of line clashing reports can now be automated through the OrcaFlex programming interface.
- The program no longer includes Rayleigh damping contributions from the geometric stiffness due to effective tension.
- The process of importing hydrodynamic data for Vessel Types has been improved in a number of ways. Now, generic text data may be combined in one file and imported in a single operation - previously, displacement RAOs, load RAOs, QTFs, and added mass and damping all had to be in separate files. Generic text data files may now specify the conventions and units of their data. The different ways of importing Vessel Type data are now consistent, whether importing all the data at once or importing a single category, i.e. RAOs, QTFs or Added Mass and Damping. In all cases, the vessel type conventions are preserved. If the conventions or units of the data being imported differ to those of the Vessel Type, then the data will be converted to match them at the time of import. Examples and descriptions are given of how to mark up RAO, QTF, and added mass and damping output from various analysis packages for import into OrcaFlex.
- Vessel QTFs are now dimensional - previously they were non-dimensional. Old binary (.dat) data files containing non-dimensional values will be converted automatically when read. For text data files from old programs, the non-dimensional data will be read in but not re-dimensionalised.
- First order displacement and load RAOs now handle extrapolation towards zero period response differently. In previous versions if any extrapolation was needed you were warned about it but the program extrapolated based on the first two rows of the RAO tables. This meant that it would not necessarily extrapolate towards a zero response. In this version, if extrapolation is needed, OrcaFlex will now extrapolate towards a zero response for zero period. You are still warned that the data is missing and that the program will do this extrapolation for you. If you specify non-zero data for zero period then this data will be replaced with zero response.
- For primary motion time histories, a degree of freedom set to None now fixes the degree of freedom as the value at the end of the static calculation. Previously this setting fixed the value as zero.
- Vessel time history data now includes an origin which specifies the point on the vessel at which the motion time history applies.
- Generation of vessel response reports can now be automated through the batch script, or from the OrcaFlex programming interface.
Elastic solid contact modelling
Elastic solid contact is now handled slightly differently for objects with non-zero contact diameter, e.g. line-solid contact. The algorithm used in previous versions had some weaknesses. The essential problem was that the reaction force always acted normally to the solid face closest to the node. When a node is near an edge or a corner then this results in a discontinuity which in turn can lead to poor convergence and noisy responses.
In addition, the previous algorithm had the unfortunate property that contact could, under some circumstances, occur for nodes outside the shape. This problem most commonly afflicted nodes just beyond the end of a curved plate, e.g. a bellmouth.
This version of OrcaFlex solves these problems by adopting a minor change to the contact algorithm. The reaction force now acts in the direction of the vector po-ps where po is the position of the node and ps the position of the closest point on the surface of the solid to po. When ps is on one of the faces of the shape then the new and old algorithms are identical. However, when ps is on an edge or at a corner then the reaction force is no longer normal to one of the faces.
Should you require more technical details regarding this change, please contact Orcina.
A new type of shape, the drawing shape, is available for drawing purposes. Drawing shapes have no physical effect on the model and do not interact with other objects.
- A new eigen-solver has been implemented to perform modal analysis. The new solver is much faster and requires much less memory than the previous solver and this has significant benefits for large problems. The new solver is not applicable to all problems and so the old solver is retained as a fall-back. More details are given in the modal analysis theory topic.
- Modal analysis was previously only available for individual lines. It is now possible to perform a modal analysis for the whole system.
- Modal analysis is now available for systems with non-banded stiffness matrices. For example, previous versions of the program would not perform a modal analysis if there was a link connecting two nodes on the same line but modal analyses for such models are now possible.
- The VIV table on the modal analysis form previously allowed you to control how the modes were output by sorting the modes on offset distribution. This was not the most helpful user interface because typically you wish to filter out less interesting mode types (e.g. only show transverse modes) and display in increasing mode number order. The new filter option implements this improved user interface.
- There is a new option to collate, sum and present fatigue damage output from SHEAR7.
- Another new option is available for mooring fatigue where the damage calculation is based on T-N curves.
- Mean stress effects can now be accounted for using Goodman, Soderberg or Gerber models.
- Stress factor fatigue now offers the choice of wall tension or effective tension as the tension variable. Previous versions always used wall tension.
- A new menu item, View | Show Entire Model, has been added which sets the view parameters for the active 3D View so that the entire model will be displayed. Related to this change, new data has been added to the drawing page on the General data form to allow more control over default view parameters. Previous versions of the program stored default view parameters as absolute, fixed values. The new option allows the default view to be determined by the extent of the model.
- Copy/paste and import from Library files for objects which refer to variable data has been improved. In previous versions the variable data sources were not imported which meant that you had to separately paste or import the required variable data items. In this version all variable data sources that are referenced by the imported objects are also imported. This means that variable data references are now handled in the same way as type objects (e.g. Vessel Types, Line Types).
- Shaded graphics video export now offers direct support for the Windows Media Video 9 codec, which is identified by the code WMV3. This codec produces lower quality videos than XVID for the same video file size, but does have the advantage that the videos should work on almost all Windows machines without the need for codec installation.
- A new menu item, File | Properties, has been added which displays the system file properties dialog for the current file. This has been added to make it easier to find the full path for files with long names.
- GIF and JPG images can now be exported from 3D View and Graph windows.
- The OrcaFlex Spreadsheet has been completely re-written. Previous versions were implemented using VBA but the new version removes all VBA and instead implements the functionality as a COM add-in to Excel. From the user's perspective this change is largely immaterial - the operation of the spreadsheet remains essentially unchanged. The main advantage of, and indeed the principal motivation for, the change, is that the spreadsheet can now make use of multi-core and multi-processor hardware to post-process multiple load cases in parallel.
- The Linked Statistics post-processing instruction now reports m0, m2, m4, Tz, Tc and spectral bandwidth parameter e. This brings it into line with the Linked Statistics output produced by the interactive OrcaFlex interface.
- The Duplicate Instructions action now duplicates cells by using Excel's standard copy routines. This means that formulae will be copied, and references updated in the standard Excel fashion. A common request has been to allow an easy way to output the load case file name to an output sheet. The obvious way to implement this is with a label defined by a formula referring to the load case file name. This approach failed in previous versions but this change, by allowing formulae to be copied in the standard Excel fashion, solves the problem.
- A new facility has been added to automate generation of text data files. This is very similar to the existing facility for generating batch script files.
- Hidden rows in post-processing instructions tables are now ignored. This allows you to disable certain instructions which can be useful when developing and building a table of post-processing instructions.
- Post-processing can now take advantage of Excel 2007 and later supporting very large worksheets. Older versions of Excel were limited to 65536 rows and 256 columns, but the more recent versions of Excel have effectively removed these limits.
OrcaFlex Programming Interface
- We now provide an interface to OrcaFlex from MATLAB. This MATLAB interface is functionally identical to the Python interface.
- Wave search output, summary results and full results can now be exported from the API.
Network dongle config files
The preferred location for the network dongle configuration file, NetHasp.ini, is now COMMONAPPDATA\Orcina. The OrcaFlex.msi installer will, if required, write the file to this location (for a new installation) or copy it from its old location (for an upgrade). See the OrcaDongle help file for further details.
- Ramping during build-up for vessel primary motion time histories was handled incorrectly when 6 DOF statics was in use.
- Simulations using time domain VIV, for models which included bend stiffeners, would sometimes become unstable.
- Cancelling data forms was, in some situations, failing to force a refresh of shaded graphics meshes.
- Previous versions were not capable of opening text data files as library files.
- OrcFxAPI was failing to load certain text data files when specified by a relative file name.
- Modal analysis views with Draw node axes enabled for lines with attached links or winches was resulting in a corruption of static state positions for such links or winches. This corruption could also affect the first log sample of simulations. As a workaround, reset the model and repeat the static calculation.
- End load results for free-ended lines with attached stiffeners erroneously included the stiffener contribution to end load.
These bugs are all fixed in version 9.4a.