Mechanical concepts, roughed.
Also called: Mechanical architecture · Concept layout · Structural concept · Pre-CAD massing
Roughing out how the product is built: its structure, how parts fit and are made, where the heat and insulation go, before any detailed CAD.
A mechanical concept is the structural plan before CAD: walls, materials, the heat path, insulation, how it goes together. Get the physics right on paper and detailed design is cheap. Get it wrong and CAD just renders the mistake in high resolution.
What mechanical concepts are
A mechanical concept is the structural idea of the product before you draw a single dimensioned part. It answers the load-bearing questions: what is the body made of, how thick are the walls, where does heat come from and where does it go, what insulates what, how do the pieces locate against each other, and how is each part actually made. It is the bridge between “we want a heated box” and a CAD model that someone can tool.
The point of doing it on paper, or on a whiteboard, is that the expensive decisions are the early ones. Wall section, material choice, and heat path set the cost, the thermal performance, and the manufacturing route. Change them after the CAD is detailed and you redo the CAD. Change them on a napkin and you redo a napkin. I have watched teams skip this and spend three weeks modelling a structure that physics was never going to allow.
A good concept usually carries two or three viable options, not one. You sketch the obvious approach, then a cheaper one and a more robust one, and you compare them against the target the product has to hit. For the proofing box that target was concrete: hold 26°C ±0.5°C at under 30W, in a UK kitchen that swings between 14°C and 19°C overnight. Every structural decision was checked against that one line.
The concept that wins is the one that hits the thermal target, not the one that looks neatest in a render. Here is that distinction set out plainly.
- Lock the shape and finish first, then ask where the heat goes.
- Pick a thin, light shell because it looks elegant in CAD.
- Treat insulation and the air gap as something to add later.
- Discover at prototype that it can’t hold 26°C and restyle around the physics.
- Start from the target: 26°C ±0.5°C at under 30W.
- Choose double-wall ceramic mass because the physics needs it.
- Design the air-gap insulation in from the first sketch.
- Let the wood band and finish wrap a structure that already works.
The left column is how good-looking products fail thermal testing. The right column is slower to feel finished, but it is the only one where the prototype behaves the way the brochure promises.
How it fits the bigger picture
Mechanical concepts is an early Stage 06 Design activity. It takes the design brief and the defined scope and turns them into a structural plan, which then feeds straight into electronic concepts (06.10.05), where the heating element, the sensor, and the control board are worked out against the structure you have just set.
What it can do
It commits the expensive decisions, walls, materials, heat path, and insulation, on paper where they are cheap to change. It gives the electronics and detailed CAD a structure to build inside, and it surfaces the killer trade-offs (mass versus cost, seal versus ease of use) while they are still arguments rather than tooling invoices.
What it can’t do
It can’t prove the numbers. A concept that says “the air gap will hold it under 30W” is a hypothesis until a thermal model and a prototype say so. It also can’t fix a vague brief; if the target isn’t a hard number like 26°C ±0.5°C, the concept has nothing to be checked against.
See the full 10-stage process →
Try it yourself
Take your product’s one hard performance target and write it at the top of a page. Then sketch the structure that delivers it: walls and their thickness, what each part is made of and how, where heat or load enters and leaves, what insulates or isolates what, and how the pieces locate. Do it twice more, cheaper and more robust, and hold all three against the target. The one that hits the number wins, even if it isn’t the prettiest.
Want a guided start before you sketch? Start the Free Sprint → and the GPT will help you pin the target the concept has to hit.
Your mechanical-concept checklist
Project notes: ceramic mass over a clever shell
▸ From the notebook · optional reading
Project notes: the wall section that decided everything
Three concepts on the whiteboard with Dan and Anna Hartley in Stockport. The thin clever shell lost to the heavy ceramic one, and the air gap is why.
3 min read · click to open
We started the design stage with the target pinned on the wall: hold 26°C, give or take half a degree, at under 30W, in a Stockport kitchen that drops to 14°C overnight. I asked the room to give me three structural concepts before anyone opened CAD.
The three on the whiteboard
Concept A, the thin shell. A single moulded wall, light, cheap, easy to make, with the element bonded to the base. Dan liked it on cost. The trouble showed up the moment we sketched the heat path: with almost no thermal mass, the inside air would track the element switching on and off, and a 14°C kitchen would pull heat straight back out through the thin wall. Holding ±0.5°C would have meant a fast-cycling controller fighting the physics all night.
Concept B, the double-wall ceramic. Two ceramic walls with an air gap between them, the element in the base, a wood band on the outside. Heavier, dearer, slower to warm up. But the mass smooths the temperature and the air gap keeps the heat in, so the load stays under 30W and the outer wall stays cool enough to touch. This is the one we carried forward.
Concept C, ceramic with a foam-insulated jacket. Better still on paper for insulation, but it lost the look entirely and added a part that had no business near a humid, warm interior. Anna killed it on the spot: it stopped being something you would leave on a worktop.
Why the wall section won the argument
The whole decision came down to wall section. The air gap between the two ceramic walls was doing the insulating job a foam jacket would have done, without the extra part and without spoiling the object. Once we drew that gap, the ±0.5°C target stopped being a fight and became a consequence of the structure. We worked the numbers roughly on the whiteboard, enough to be confident the 30W budget held, and left the proper thermal model for the next activity.
It cost us: the ceramic mass added weight and a slow warm-up, and the Stoke-on-Trent tooling was the most expensive single line on the bill of materials, which lands the unit between £38 and £55. We took that hit deliberately. The mass is the reason the box holds temperature, and holding temperature is the entire product. The lid stayed a simple lift-off that seals enough to hold humidity, because a latch is one more thing to fight at six in the morning. UKCA and BS EN 61010 framed the element and isolation choices from the first sketch, not as an afterthought.
— Design stage, project notes, 2026
— Next in Design → Electronic concepts