General

General

What does the error: “Unable to compute interface connectivity between the following domain interfaces” mean when running a Fourier Transformation Transient Blade Row case?

    • FAQFAQ
      Participant

      What does the error: “Unable to compute interface connectivity between the following domain interfaces” mean when running a Fourier Transformation Transient Blade Row case? ################################################################## Running Fourier Transformation Transient Blade Row cases with single precision in both the partitioning and solver will lead to the error shown below: +——————————————————————–+ | ERROR #999100267 has occurred in subroutine gKCsFCCsPS. | | Message: | | | | Unable to compute interface connectivity between the following | | domain interfaces: | | | | Phase Corrected domain interface : OGV to OGV Periodic 1 | | Sampling domain interface : Sample OGV | | | | for this serial/partitioning run. Likely causes are: | | | | 1) The mesh has been transformed using single precision | | arithmetic at some point. | | 2) The single precision solver/partitioner is being used instead | | of the double precision solver/partitioner as recommended for | | the Fourier Transformation model. | | | | If 2) is not the case please increase the expert parameter | | “ps mapping check tolerance” to a value larger than the current | | value: 1.000E-04. | | | +——————————————————————–+ This is related to the mapping between the sampling interface (where the Fourier coefficients are collected) and the phase shifted interface where the Fourier coefficients are used to reconstruct a phase shifted signal. This mapping is done numerically where the solver looks for matching control surfaces between the two interfaces (the interfaces could either be 1:1 or non-conformal). The tolerance is set very low, requiring the use of running either the solver or the partitioner in double precision to avoid errors in phase shifting. In parallel, the mapping is done in the partitioning phase of the run and saved to the par file, so one possible work flow is to run the partitioner in double precision and the solver in single precision. Therefore, to avoid this error message, select one of the run settings below: 1)Run both the partitioner and solver in double precision. 2)Run the partitioner in double precision. Run the solver in single precision 3) Run the partitioner in single precision. Run the solver in double precision 4) For a serial run, the solver must be run in double precision If the error is still being generated even after trying to run in double precision, check the setup in CFX-Pre. If multiple periodic interfaces or sampling interfaces exist for a given blade row, this will trigger the same error message. Multiple periodic or sampling interfaces can result from generating the mesh in TurboGrid and having separate inlet and/or outlet blocks generated. Separate periodic and sample interfaces can be defined on the inlet, passage and outlet block. However, the transient blade row setup only allows for one of these interfaces to be selected. Running the solver will then generate the error. To avoid this, define a single periodic and sample interface for each blade row.