fix bug in bldprcnt following issue #3416

[ptrbortolotti]

Work on issue #3416

Thank you @Sparsh-Sharma for the detailed issue. You are very right that the code breaks for BldPrcnt=100. Surprisingly enough, we never set it to 100%, and we never noticed the problem.

I've adopted your suggestion of setting p%startnode = min(p%NumBlNds, 2) instead of p%startnode = max(1, p%NumBlNds - 1).

Next, I think I am good to remove the check p%startnode = max(min(p%NumBlNds,2),p%startnode), which should never be triggered

Lastly, I added a warning if BldPrcnt is smaller than the last element size. I think that's the warning you were recommending. You wrote 'A warning if BldPrcnt is large enough that no node satisfies the threshold might also be worth adding, since the condition is silent today.', but I was a little lost. I think this happens when BldPrcnt is too small, not 'large enough'. Let me know if I am still missing something.

I will also improve the documentation in the r-tests now

[Sparsh-Sharma]

@ptrbortolotti
👍

[Copilot]

Pull request overview

Fixes an edge-case in AeroDyn’s AeroAcoustics span-selection logic so that BldPrcnt=100 correctly uses the full blade span (instead of falling back to only the outermost nodes), aligning behavior with the documented meaning and the reported bug in #3416.

Changes:

• Changes the startnode fallback initialization to min(NumBlNds, 2) so “no node satisfies threshold” correctly degrades to “use the whole blade”.
• Removes the now-redundant post-loop clamp of startnode.
• Adds a warning when BldPrcnt is smaller than the last blade element size (i.e., when the minimum-possible span selection is implicitly used).

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.