Module 20
Checking, review and knowing what you do not know
The question the other nineteen modules assumed an answer to: how do you find out whether a design is right — and how does a design say what it cannot vouch for?
What this module covers
- Check an answer by order of magnitude before checking it by arithmetic
- Say what each level of independence can find, and what it is structurally blind to
- Rank findings by consequence rather than by the size of the numerical error
- Ask the review questions that find the most per minute spent
- Read this course's own register of unverified values, and say why it is generated
- State honestly what this course is and is not fit for
Lessons
Not by reading the calculation again. This lesson is about the three things that find most of what is findable, and about why checking has levels rather than a single state called 'checked'.
Start lesson →The last lesson, and the one that turns the method back on the course itself. Every value it could not verify is listed — by the code, not by hand.
Start lesson →
Module checkpoint
Check what you have taken in
5 questions
Question 1
A beam check was at 94% of resistance. A review finds the moment was under-estimated by 9%. What is the utilisation once corrected, as a percentage?
Question 2
A brace check was at 25%. A review finds its force was under-estimated by 60%. What is the utilisation once corrected, as a percentage?
Question 3
A steel tonnage is reported as 62 000 kg/m² for a multi-storey office, where the usual range runs to 90 kg/m². By what factor does it exceed the top of the band?
Question 4
Of the 47 design values in the profile, 33 come from BS EN 1993-1-1 alone. What proportion of the profile is that, as a percentage?
Question 5
A frame is reported with αcr = 3.75. The usual range for a braced frame starts at 10. By what factor does it fall below the band?