ENH: parameterization of hopper blockMeshDict

I've parameterized the blockMeshDict for the hopper simulations for a own case using the variables and the modern #eval function. This makes the blockMeshDict much more readable and could serve as a template for users writing blockMeshDicts. Would it be possible to push this the develop version? I can imagine it is a hectic time during the release. Just let this sit until afterwards.blockMeshDict

changed the description

Hi @bjnieuwboer,

Thank you very much for your contribution.

Would you mind to tell us which tutorial's blockMeshDict was aimed for? lagrangian/icoUncoupledKinematicParcelFoam/hopper?

Yes, indeed. it is for /lagrangian/icoUncoupledKinematicParcelFoam/hopper/hopperEmptying

I just realized that the hopperInitialState has a different mesh. I will create a blockMesh for that one tomorrow and attach it.

This is the blockMesh for the lagrangian/icoUncoupledKinematicParcelFoam/hopper/hopperInitialState: blockMeshDict

assigned to @kuti

added contribution label

added community label

mentioned in commit 346890c5

mentioned in commit 7123e7983881131a095e38748505a306f4cde0e1

mentioned in commit f7920a31bf5af4b881c01ecf21e96d31b311cd30

mentioned in commit 8f6af90c

mentioned in merge request !476 (merged)

mentioned in commit 175c1849

Dear @bjnieuwboer

I have observed that the contributions of yours led to small differences in polyMesh, which in turn led to small differences in flow predictions.

Could you please double check the contribution items and verify that they do not cause any changes on anything (unless necessary) by performing elementwise comparisons between old and new states of output files (e.g. between old and new polyMesh/points)?

Many thanks.

Dear @kuti,

Thanks for double checking my contribution. I am happy that you take this precaution. I just did a check for the changes for both cases and attached the results. pointsInitialOrig and pointsInitialStateParameter show the points file for the hopperInitialState-case. Where the latter shows the points file using the new contribution. Similar for the pointsEmptyingOrig and pointsEmptyingParameter for the hopperEmptying-case.

When comparing the files it shows a round-off error after the 7th decimal. I can imagine that this will lead to slightly different flow results. However, I suppose that the geometry is better represented using the parameterized blockmesh file, since it is likely that the bottom angle of the hopper was supposed to be 30 degrees and not slightly more or less.

pointsInitialStateOrig

pointsInitialStateParameter

pointsEmptyingOrig

pointsEmptyingParameter

mentioned in commit 4afdfe75

mentioned in merge request !482 (merged)

mentioned in commit c64c312b

@bjnieuwboer , highly appreciated - now merged. many thanks.

closed

mentioned in commit 8930b314