The Design Process: From Sketch to CAD to Reality
There's a time-honored saying about creating something:
First you make it.
Then you make it right.
Then you make it better.
You inevitably learn so much from the first build. And the second. And the third.
The Napkin Sketch Phase
Every product starts with a sketch. Mine started on the back of a receipt at a coffee shop.
A circle. Then another circle inside it. Some lines indicating panels. A note: "fits in parking spot?"
That sketch sat in my notebook for months. But the idea wouldn't let go. What if we really could create a modular, mobile growing space that fit in a parking spot?
Moving to 3D Modeling
The first step was to work out the idea in a 3D modeling program. I used Blender—free, powerful, and perfect for exploring forms without worrying about real-world constraints yet.
In Blender, I could:
- Visualize the sphere in three dimensions
- Experiment with panel arrangements (how many? what shapes?)
- Test different configurations (doors, windows, vents)
- Create renderings to communicate the concept
This phase is about exploration. What does it look like? How does it feel? Does the form communicate the function?
I tried dozens of variations:
- Different panel counts (12, 22, 32, 42)
- Different elevations (on the ground, on legs, on a platform)
- Different opening mechanisms (hinged panels, sliding panels, removable panels)
The 22-panel geodesic configuration emerged as the sweet spot. Enough panels for flexibility, few enough for simplicity.
The CAD Phase: Making It Real
3D modeling is art. CAD is engineering.
Moving from Blender to CAD (I use OnShape for its collaboration features and cloud-based workflow) meant confronting reality:
- Actual dimensions: Not "about 8 feet," but exactly 8 feet 2 inches
- Material thicknesses: 3/4" plywood, 2x4 framing, specific hardware
- Tolerances: How much gap between panels? How tight should joints be?
- Assembly sequence: What order do the panels go on?
- Structural calculations: Will this actually stand up?
The Constraints
CAD forces you to make decisions:
Size: We settled on a diameter that allows two units in one parking spot (9' x 18'). This meant an 8-foot outer diameter, which also fits through a standard garage door.
Materials: Standard lumber sizes (2x4s, 2x6s) and sheet goods (4x8 plywood). Using standard materials means:
- Available everywhere
- Affordable
- Familiar to builders
- Replaceable if damaged
Hardware: Off-the-shelf connectors, hinges, and fasteners. No custom machined parts. If someone in rural Montana wants to build one, they can get everything at their local hardware store.
Tools: Buildable with common tools—circular saw, drill, basic hand tools. No CNC router required (though it helps). No specialized equipment.
The Bill of Materials
One of the most valuable outputs of the CAD phase is the Bill of Materials (BoM).
For the Thiosphere, this includes:
- Exact lumber quantities (number of 2x4s, cut lengths)
- Plywood sheet count and cutting diagrams
- Hardware list (screws, bolts, hinges, connectors)
- Optional components (windows, vents, doors)
- Estimated costs
The BoM makes the project real. It's no longer "build a sphere." It's "buy these specific materials and assemble them in this specific way."
Prototype Zero: The Alpha Build
With CAD files in hand, it was time to build.
The first prototype—I call it Proto 0—was built in my garage. The goals were simple:
- Prove the concept works
- Identify what we got wrong
- Learn what we didn't know we didn't know
What Went Right
- The sphere assembled (not a given!)
- The structure was stable (no wobbling or sagging)
- The nested sphere concept worked (thermal performance was impressive)
- The form factor was correct (it fit where we expected)
What Went Wrong
- Panel alignment was tricky (tolerances were too tight in some places, too loose in others)
- Assembly sequence mattered (we had to disassemble and reassemble twice to get it right)
- Some joints were over-engineered (too many fasteners, too complex)
- Weather sealing was an afterthought (it leaked in the first rain)
What We Learned
The big lessons from Proto 0:
-
Tolerances matter: We needed to design in adjustment. Panels should have 1/8" gaps, not press-fit tight.
-
Assembly sequence is critical: The last panel is the hardest to install. We needed to design the sequence from the beginning.
-
Simplify joints: Our initial design had complex angle cuts. We redesigned for simple 90-degree cuts wherever possible.
-
Weather sealing from the start: Seals and gaskets aren't optional. They need to be designed in, not added later.
-
Thermal performance exceeded expectations: Even without insulation, the nested sphere maintained temperature remarkably well.
The Iteration Process
After Proto 0, we went back to CAD. Every lesson learned was incorporated:
- Revised panel dimensions (added tolerance gaps)
- Simplified joints (fewer angle cuts, more standardized connections)
- Integrated weather sealing (gasket channels designed into panels)
- Improved assembly sequence (documented step-by-step)
- Modular configurations (designed different panel types for different uses)
This is the "make it right" phase. We know it works. Now we make it work well.
Prototype One: The Shadosphere → Saunosphere
For Proto 1, we decided to build the simplest possible configuration: a sauna.
Why a sauna?
- Simple requirements: Heat, ventilation, seating
- Rustic aesthetic: Imperfections are charming, not problematic
- Immediate value: People understand saunas
- Forgiving use case: Small variations in temperature are fine
The Saunosphere (we started calling it the Shadosphere for shade/shadow, but Saunosphere stuck) taught us:
Material Choices Matter
We used cedar for the interior—traditional for saunas, naturally resistant to moisture and heat. But we learned that not all cedar is equal. Clear cedar is expensive. Knotty cedar is affordable but requires careful selection.
Ventilation Is Critical
The convection properties of the sphere are excellent, but you still need designed ventilation. We added:
- A vent at the top (hot air exhaust)
- A vent at the bottom (cool air intake)
- Adjustable louvers (control airflow)
Heat Source Integration
We tested both electric and wood-fired sauna stoves. The sphere's geometry meant we could use a smaller stove than a rectangular sauna of the same volume. The even heat distribution meant no cold corners.
User Experience
This was the first time we had real users. Friends, family, neighbors—everyone wanted to try the sauna sphere.
Feedback:
- "It feels bigger inside than it looks outside" (the sphere's geometry)
- "The heat is really even" (convection working as designed)
- "It's beautiful" (aesthetic validation)
- "Can I buy one?" (market validation!)
Prototype Two: Increasing Complexity
Proto 2 is planned as an Immosphere—an immersive video space with screens, power, and climate control.
This adds:
- Electrical systems (power distribution, outlets, lighting)
- HVAC requirements (heat pump or AC unit)
- Higher fit and finish (visible imperfections matter more)
- Technology integration (screens, audio, controls)
We're currently in the design phase for Proto 2, incorporating all the lessons from Proto 0 and Proto 1.
Prototype Three: The Agrosphere
Proto 3 will be the original vision—a mobile greenhouse.
This is the most complex configuration:
- Water management (irrigation, drainage, humidity control)
- Plant support systems (shelving, grow lights, timers)
- Environmental monitoring (temperature, humidity, light sensors)
- Automation potential (smart controls, remote monitoring)
We're deliberately building up to this complexity. Each prototype teaches us lessons that apply to the next.
The Open Source Commitment
Throughout this process, we're documenting everything:
- CAD files (OnShape models, available on GitHub)
- Build instructions (step-by-step with photos)
- Lessons learned (what worked, what didn't, what we'd change)
- Bill of materials (exact parts and costs)
- Assembly videos (visual guides for builders)
All of this will be released under the CERN Open Hardware License with Strong Reciprocity. Anyone can use these designs to build a Thiosphere for commercial purposes immediately.
Make It, Make It Right, Make It Better
We're currently in the "make it right" phase. Proto 0 proved the concept. Proto 1 is refining the execution. Proto 2 and 3 will expand the capabilities.
But the process never really ends. Even after we release the designs, the community will improve them. Someone will find a better joint. Someone will optimize the panel layout. Someone will discover a new configuration we never imagined.
That's the power of open source. We make it. The community makes it better.
And it all started with a sketch on the back of a receipt.
Next Post: Zero-K Farming: Growing Food with Minimal Capital
Previous Post: The Shape: Why Spheres Make Sense
Related: