• Skip to main content
  • Skip to search
  • Skip to footer
Cadence Home
  • This search text may be transcribed, used, stored, or accessed by our third-party service providers per our Cookie Policy and Privacy Policy.

  1. Community Forums
  2. Optimization
  3. Autoblade: Number of slices in ''Build blade sections from...

Stats

  • Replies 4
  • Subscribers 5
  • Views 3859
  • Members are here 0
More Content

Autoblade: Number of slices in ''Build blade sections from CAD surfaces'' affecting surface generation

JosephSmith
JosephSmith over 3 years ago

Hi all,

I had meshed a centrifugal impeller in Autogrid, and intended to use that geomturbo file as the target model for parametrization in Autoblade.

The target model did not have sectional data and I had to use the ''Build blade sections from CAD surfaces'' to make sliced sections. First, I chose two sections at 0.01 and 0.99. I used the compressor template as the initial parametric template and arrived at a reasonably accurate version of the blade, seen in the 3 images:

Albeit the curvature at the leading edge was not as accurate seen at the B2B view of section 1:

I tabulated my selected parameters below:

My problem is when I chose more than two sections for the initial slicing of the target model in the ''Build blade sections from CAD surfaces'' option: When I tried with 3 sections at 0.01 0.5 0.99, I got a drastically curved leading edge and unusual trailing edge:





The blade to blade view for the middle section shows that the leading edge stacking point is drastically far off from target geometry:



I am uncertain as to why this is the case. I have been tinkering with the Stacking Laws options but have not had luck identifying how this issue occurred or ways to resolve it. I would be grateful for a second opinion on this situation

  • Sign in to reply
  • Cancel
  • domen
    domen over 3 years ago

    Hi,

    Quick question: are the last images taken after "fitting" the blade, or just after adding the cuts? In your parametrization, the blade is defined along dm/R - theta, with stacking at the leading edge. Even a small difference of theta at the beginning may generate a big "distortion" for a centrifugal blade at the trailing edge. It should be possible fixing it, either by "fitting" the blade or by adjusting the sections manually.

    Thanks,

    • Cancel
    • Vote Up +1 Vote Down
    • Sign in to reply
    • Cancel
  • JosephSmith
    JosephSmith over 3 years ago in reply to domen

    Hi Domen,

    The last two images were taken just after adding the cuts. Upon looking at the blade to blade view, I adjusted the tangential law from a line to b-spline and got a better result seen below. My mistake earlier was also not pressing the adapt button after adjusting the tangential law curves, I assumed adjusting it in the Model editor section automatically updates the view:

      



    However, when I sliced the sections 8 times and used 8 stream surfaces, I got quite distorted shapes once again even after using different tangential laws for the lean curve. I am not yet why the addition of more stream surfaces is causing the unusual shape

    • Cancel
    • Vote Up 0 Vote Down
    • Sign in to reply
    • Cancel
  • domen
    domen over 3 years ago in reply to JosephSmith

    Missed this message, sorry!

    I don't remember if AutoBlade copies or interpolates the newly added sections from the existing ones.  Anyway, and I'm thinking about the issue in the optimization, I would use only two sections: one at the hub or just below (1% below), the other at the shroud (or just above).

    Having a lot of sections is great for a perfect fit of the blade surfaces, but the model becomes impossible (for a human at least) to set up in an optimization. Having 8 independent sections is going to create a really weird looking (and probably impossible to manufacture) blade.

    Since you want to optimize this geometry, I wouldn't worry too much about accuracy: clearly, the geometry is going to change, as long as the new baseline design isn't performing worse than the original one, the optimizer will give you something new.

    My advice is to get the parametrization right, not the actual values of the parameters, which are just an output of the problem.

    • Cancel
    • Vote Up +1 Vote Down
    • Sign in to reply
    • Cancel
  • JosephSmith
    JosephSmith over 3 years ago in reply to domen

    Hi Domen,

    No worries! Thanks for the insight in designing for manufacture and ultimately the optimization process in the long run. I will redo it with two sections. I intended on trying only with one free parameter and single optimization to get comfortable with the process, and then extend it more free parameters and multiobjective optimization.

    • Cancel
    • Vote Up 0 Vote Down
    • Sign in to reply
    • Cancel
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.

© 2025 Cadence Design Systems, Inc. All Rights Reserved.

  • Terms of Use
  • Privacy
  • Cookie Policy
  • US Trademarks
  • Do Not Sell or Share My Personal Information