The Sandwich Must Hold

Weight 40% · Capability: Synthesise · Due 28 May 2027, end of Week 12

Design, test, revise and defend one sandwich system against a load case you declare in advance.

Task at a glance#

• Mode — individual.
• Format — a design and test dossier: load-case brief, exploded diagram, declared constraints, protocol, evidence, one documented failure, one justified revision, and a final structural defence.
• Scale — one system, one load case, at least two rounds of testing with a revision between them.
• Load case — choose one: backpack transport, one-handed use, delayed service, vertical storage, or repeated handling.
• Evidence — physical specimens, non-food analogues, supplied evidence, or any combination.
• Route — one file, submitted to the SLOP2895 submission point before the deadline.

The brief#

The three earlier tasks each held one part of the method still. The Field Log observed structures without changing them; the Component Test changed one variable and held the rest; the Failure Autopsy worked backwards from a failure someone else had documented. This task runs the whole loop forwards, on a system you are responsible for, against a condition you commit to before you know whether your design survives it.

The declared problem is that structural performance is meaningless without a stated load case. A system is not strong; it is strong against something. An assembly that holds perfectly on a bench may not survive forty minutes upright in a bag, and one built for that journey may be unusable in one hand. Declaring the load case first is what makes the rest of the dossier arguable: it fixes what counts as success, what counts as failure, and which trade-offs you were entitled to make.

The iteration record is the core of this submission. You are required to document a failure, and to make one revision that answers it, with evidence on both sides of the change. This is not a penalty clause. A design that never failed was tested too gently, and a dossier with no failure in it will be read as a dossier with no test in it. A structure that holds only when it is carried carefully has not been tested; it has been escorted.

Taste, visual appeal, culinary skill, ingredient expense and whether the result is eaten are explicitly excluded from marking. An attractive system with no load case, no protocol and no evidence marks low. An unattractive one with a declared load case, a repeatable test, an honest failure and a defensible revision marks high.

Required submission#

1. A load-case brief: the case chosen, the conditions it imposes, the duration or repetitions involved, and the criteria that separate holding from failing.
2. An annotated exploded diagram of the system, showing every layer, every interface and the load path through the assembly.
3. Declared design constraints, including the ones imposed by the load case and the ones you accepted for cost, waste or accessibility reasons.
4. A test protocol capable of applying the declared load case repeatably.
5. Test evidence from every round, keyed to the protocol.
6. One documented failure, located at the interface or layer where it began.
7. One justified revision, stating the failure it answers, what changed, what was deliberately left unchanged, and the evidence after the change.
8. A final structural defence.

Recommended process#

1. Declare the load case and the criteria for holding and failing, and date the declaration before the first test.
2. Classify the layers of your system by function — structural, interface, or carried load — and mark the load path on the diagram.
3. Document the constraints you are designing within before you design within them.
4. Test the system against the declared case, in full, without moderating the conditions to protect it.
5. Observe and measure the failure where it initiates rather than where it becomes obvious.
6. Diagnose the failure using the causal discipline of the Autopsy: evidence, sequence, and at least one alternative you ruled out.
7. Revise one thing, deliberately, and record what you held constant so the second round remains a comparison.
8. Compare the two rounds under the identical protocol.
9. Defend the final system by stating what it survives, what it does not, and the evidence for each.

Evidence standard#

The dossier must let a marker separate four things at a glance.

• Observed

  Drawn solid: present in the evidence.

• Inferred

  Drawn dashed: reasoning laid over the evidence.

• Uncertain

  Drawn as an open span: the evidence fixes no one value.

• Limitation

  Drawn as far as the method reached, and no further.

Observed is the recorded behaviour under the protocol: what deflected, what migrated, what separated, when, and at which point in the load case. Failed rounds are reported in the same detail as successful ones.

Inferred is your structural reasoning — why the load path explains the observation, why the revision should address it. Each inference names the evidence behind it.

Uncertain covers spread between repeats, conditions you could not reproduce exactly between rounds, and any result you would not stake the defence on.

Limitation is the boundary of the claim. The defence must state the conditions under which the system is not claimed to hold, including the load cases you did not test. A defence that claims general robustness from one load case overstates its evidence and is marked as overstating it.

Equivalent participation#

Three routes are available, and all three assess the same outcome:

• Physical specimen — build and test a real assembly, with no requirement to consume it.
• Non-food analogue — build the system in foam, sponge, cardboard, clay or textile, declaring how each analogue stands in for the layer it represents. Repeated destructive testing against a load case is often better served this way.
• Supplied evidence — design against the supplied specimen and measurement packs, using their documented conditions as your test rounds.

No route loses marks, and the same criteria apply to all three. The load case, the protocol, the failure and the revision are assessable in every route. See the participation policies.

Safety, cost and waste#

No student is required to prepare, purchase, handle or consume food to complete this task, and no assessment here depends on specialist or expensive materials. Your protocol must justify the material each round consumes; minimising it is a design constraint you may declare and defend, not a bonus mark. Tested material must never be presented as safe to eat, and cutting must use supplied pre-cut specimens, diagrams or supervised equipment. Full detail is on the policies page.

Preparation#

The whole semester prepares this task, and the final three teaching weeks prepare it directly.

• Week 10: Packaging Is Part of the Structure — transport as a declared load case rather than an afterthought.
• Week 11: Failure Autopsy — the diagnostic discipline your documented failure is held to.
• Week 12: The Final Stress Test — whether a design stays coherent under a declared use condition.

The three earlier assessments supply the components: observation from the Sandwich Field Log, isolation and protocol from the Component Test, and causal reasoning from the Failure Autopsy.