Turn a need into testable design requirements
Write measurable requirements and distinguish a feasible model candidate from a physically validated design.
Engineering design and scientific computing
Turn a need into a measurable requirement, compare designs under the same conditions and explain what your evidence supports.
This is the course outline. Open a unit to choose a lesson; every lesson opens on its own page. Explanations and experiments live inside those lesson pages.
Course scope and preparation
Start without programming. These 23 lessons use steady conduction, calibration, transient thermal comparison, limited-reading circuit diagnosis, sampled feedback control and bounded design-suite comparisons. Preserve a baseline, identify resource constraints, check units and distinguish a calculated design from a measured physical prototype. Algebra, percentages and graph reading are explained where needed.
What is available
Design foundations, matched thermal prototypes, circuit diagnosis and controller comparisons with sensor delay and actuator limits, failure/tradeoff analysis and reproducible design arguments are available below, with twelve public cumulative assessment cases. The cumulative device-model project retains a design plan, stress evidence, an assigned transfer response and first/revised reports for human review. Software availability does not establish a semester of teaching effectiveness or physical laboratory proficiency.
Return directly to steady conduction, circuit measurement or scientific inquiry. Earlier lessons are recommendations, not navigation locks.
Requirements, models and fair comparisons
Check units and boundaries before trusting a model
Calculate a two-layer conduction prediction with consistent units and identify conditions that the model does not represent.
Keep calibration evidence in a fair comparison
Apply a known additive correction, distinguish resolution from accuracy and plan a matched design comparison.
Thermal models and controlled prototypes
Compare thermal prototypes under matched conditions
Identify a matched baseline and explain a model tradeoff between energy, temperature and resource cost.
Explain changing temperature with an energy balance
Interpret heat capacity, conductance and time constant, distinguishing heated control from a separate passive-cooling example.
Distinguish a design change from calculation error
Separate model sensitivity, time-step approximation, measurement residuals and physical model inadequacy.
Preserve a failed design and justify a revision
Document a failed requirement, compare a revised candidate and distinguish simulated verification from physical validation.
Circuit diagnosis with limited readings
Predict circuit readings from competing mechanisms
Calculate healthy series values and distinguish an open path, a bypass and a changed ideal source.
Choose a discriminating reading within a budget
Plan sequential readings, retain ambiguity and distinguish a repeated lookup from new independent evidence.
Locate a fault using node and current constraints
Use junction potential and loop current to distinguish the supplied fault cases without equating zero voltage with an open circuit.
Bound a diagnosis by its hypotheses and instruments
Explain why a unique finite-case match does not exclude unmodeled faults, meter loading or calibration errors.
Sensors, feedback, delay and limits
Trace a sensor–controller–actuator loop
Explain feedforward plus proportional feedback, sign conventions and the consequences of a supplied sensor offset.
Compare gain without assuming bigger is better
Evaluate a gain change against separate time, temperature and energy evidence without claiming universal stability.
Account for the age of a feedback reading
Read sample-time receipts and explain why an old temperature can cause continued heating after the target is reached.
Distinguish actuator capacity from an energy criterion
Calculate power-limited equilibrium and energy accounting while keeping unmet temperature requirements visible.
Revise a controller while retaining its failure
Build a traceable gain revision and separate development captures from future independently reviewed transfer evidence.
Failure analysis and design tradeoffs
Turn failed requirements into a useful failure log
Reproduce a failure with exact candidate/scenario settings and retain alternatives rather than replacing the original evidence.
Distinguish scenario coverage from field reliability
Report bounded pass counts, preserve common assumptions and identify evidence needed for a reliability estimate.
Compare feasible designs before ranking preferences
Identify nondominated feasible candidates and explain why a two-objective comparison does not select a universal best design.
Plan evidence for an unfamiliar failure
Prepare a discriminating investigation and distinguish public suite development from human-reviewed unfamiliar transfer.
Reproducible analysis and design communication
Make a result reproducible from its evidence
Prepare a reproducible model record and separate implementation agreement from physical validation.
Defend a constrained design choice with evidence
Justify a candidate under stated priorities while retaining failed scenarios and separating local model evidence from wider claims.
Revise reasoning without erasing the first attempt
Use evidence-linked feedback, preserve the original report and distinguish supported practice from independently reviewed performance.
Projects, practice and evidence
Each lesson retains a prediction, settings, comparison captures and reflection. Practice checks and saved notebooks do not certify independent design skill or physical technique. Optional programming remains separate from the guided path.
Save a science backup or import one
Science work is stored locally in this browser, separately from mathematics and AI. It does not sync automatically. Use this science-only backup; this checkout's older all-subject export does not include science.
A legacy file is available only when the complete snapshot fits the 36 MB limit. For larger snapshots, prepare a complete file set below; never discard notebooks to make a smaller file.
Matching imported notebook drafts replace their current drafts. Omitted lessons and known first attempts remain. Export your current work first if you want to retain both drafts.
Complete file sets and course backups
A file set can hold more work than one science file. Keep its manifest and every part together. Each file is limited to 36 MB. This format needs this version of LearnAI or a compatible newer version; older versions can use only the legacy single-file backup above.
The set includes every saved science course and global continuation at this moment. Later edits need a new export. A conflicting tab exports its own visible work, which may differ from saved work in another tab.
0 file(s) selected. File order and filenames do not determine identity. Missing, changed or conflicting parts block the entire restore; no good part is applied by itself.
Science storage: loading. No account or remote computation is required.