Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 3 Current »

Introduction

  • Account for every flight test including and before W24 where we have clear logs for every flight test.

  • How often did we crash?

  • What are the most common causes?

  • We need to be learning from crashes so this document will hopefully tell us what we should focus on.

  • Every time someone asks me what these stats are I always do my best to guess, but I want a solid answer.

  • This analysis will need to be redone in the future as we do more flight tests!

Breakdown

Conclusions

  • Causes of every crash (including duplicates):

    • Weather caused tipping on landings (environmental )

    • Misjudged landing (controls pilot)

    • Motor Order (controls software)

    • Wingspare mechanical integrity (mechanical)

    • Aircraft stuck in mud (environmental)

    • Signal wire soldering (electrical)

    • Accidental disarm (controls pilot)

    • Mechanical Control Authority (mechanical)

    • Motor Mount Broke (mechanical)

    • Signal wire soldering (electrical)

    • Weather caused lack of control (environmental)

    • Controller configuration (controls software)

    • Electronic component failure (electrical)

    • Loose Motor Mount (mechanical)

  • Categorizing Even More (removing duplicates):

    • Environmental (3)

    • Controls (4)

    • Mechanical (4)

    • Electrical (3)

    • Total (14)

    • Fairly even split here I guess lol

  • We should be careful about a few cognitive biases when reviewing this data

    • sampling bias

      • we don’t have a lot of data here.

      • differences in airframe have a large impact on likely failure modes.

    • survivorship bias

      • Just because something doesn’t appear in the list doesn't mean it’s not a concern.

      • In my opinion we’ve had so many close calls because of motor order and poor soldering concerns, however, these are underrepresented in the failure modes observed

  • No labels