Stats

More Content

Recirculation near volute throat causing divergence in FineOpen

JosephSmith

Greetings,

I have been running a volute mesh independent study of a volute+impeller simulation, with water as the fluid - I have been increasing the cell count in hexpress and running the simulations with the same boundary conditions. Thus far, convergence has been achieved with coarse, medium and semi-medium meshes. However, I have encountered issues with the fine mesh:

Repeated mass flow residual patterns have been recurring and consequently, the simulation has not converged, seen in image 1. 


I noticed recirculation zones in the fine mesh simulation which was prior to the volute throat /tongue area, seen by the swirling velocity vectors:


Zoomed in:

A low velocity zone was located in this area seen by the light blue (the volute outlet is actually extended, but appears cut off in this screenshot):



The same boundary conditions were used from the converged coarse/medium/semi-medium meshes, therefore I suspected the issue might lie in the fine mesh as it is the only aspect that changed. Additionally, lowering the CFL number, trying outlet boundary conditions without backflow and other attempts did not yield converged results for the fine simulation. Therefore I went on to the fine mesh to adjust.

The fine mesh had an expansion ratio of 6.4  thats larger than the recommended 5, so I improved the mesh to get a new expansion ratio of 4.9 and am running the simulation again with this mesh. However, I would like to know if there is anything else I can try , given that recirculation only occurs with the fine mesh?

Kind regards,
Joseph

  • domen
    0

    Hi Joseph,

    Looking a the velocity field, I'd say that the computation is not yet converged (V > 100m/s). I'd start with saving the solution at two different iterations (peak and trough) to see where the results change and understand what could make them change. 

    A CFL number = 2 should be enough to run these tests.  

  • JosephSmith
    0 in reply to domen

    Hi Domen,

    Thanks for the suggestions. I decided to check out the mesh grid on the horizontal section through the volute one of the similar types of simulations , just with a different tongue radius. Visually, they looked quite sharp despite the mesh quality being acceptable and I suspected my repeated residuals were due to this.

    I then adjusted the initial mesh step of the Fine mesh, to have the same divisions in the z direction as the coarse mesh (the mesh that converged). I also inserted buffer layers on most of the curves and enabled the “Improve mesh quality near concave corners”  .Running this simulation did work. The difference between the initial and the fine mesh are seen in the pictures below.





    I did notice though that this solution worked most but not all the time.

    Kind regards,

    Joseph

  • domen
    0 in reply to JosephSmith

    OK, thanks for the feedback.

    I'm a little surprised that this mesh works much better than the previous one. Did you change by any chance also something around the rotor/stator interface?

  • JosephSmith
    0 in reply to domen

    Hi Domen,

    I don't recall changing anything settings around the rotor/stator interface.

    I was also suspicious of why it seemed to work with the former mentioned adjusted settings

  • domen
    +1 in reply to JosephSmith

    OK. In cases like these, we generally activate the backflow, it can happen that the diffuser stalls and brings instability at the outlet BC. Btw, are you imposing a mass flow rate or static pressure? Perhaps with one of your geometries the mass flow rate or pressure imposed cannot physically achieved, therefore the solver doesn't provide reliable information - but it doesn't diverge.

Cadence Guidelines

Community Guidelines

The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.