Rough Projection That Supports Body Weight When Sitting

How to Design a Rough Projection That Supports Body Weight When Sitting

I spent about six months last year building a single-seat bench using a rough projection workflow in Fusion 360, then 3D printing it in PETG. The project failed twice before it worked. Here is what I learned and how you should approach it if you are trying to do something similar. The core idea is projecting a scan or reference geometry of your body onto a supporting structure, then iterating on the contact points. Start with a rough mesh — a body scan from any phone-based app works fine, but you will need to clean up the noise. I used Meshmixer for the cleanup because it is fast enough and handles decimation without destroying the contact geometry. Import the cleaned mesh into your CAD environment and create a projection plane below it. The projection direction matters here. Most people project straight down, which creates a flat underside that looks clean but fails structurally. Instead, project along a vector that accounts for how weight actually transfers through the body when seated — slightly forward and outward. This changes where the material concentrates under the sit bones and thighs.

Once the projection is done, convert it into a solid using a shell or thin-wall tool. A 6mm wall thickness with internal ribbing starting at about 12mm spacing gives you a decent strength-to-weight ratio for a single person. Go below 4mm and you are cutting corners that will fail within weeks. The ribbing should follow the natural load paths from the projection area downward to your feet or mounting points. I oriented the ribs perpendicular to the seat plane at first, then switched them to a radial pattern after my first frame cracked at a joint.

A Specific Problem I Ran Into

During my second print attempt, the area directly under the left sit bone delaminated after about three days of use. The support material underneath that region was insufficient because the rough projection had created a very thin section there — the software was collapsing the geometry too much in low-contact areas. My workaround was to add a minimum thickness constraint of 8mm to the projection operation and then manually thicken that zone using a surface offset before converting to solid. It added about 200 grams to the final part, which is negligible, but it stopped the cracking immediately. If you are working in a different CAD tool, look for a feature called "minimum wall thickness" or "designer constraints" during the projection step. Fusion 360's 3D Print workspace has this built in. SolidWorks users can approximate it with the Shell feature followed by a manual check using the Measure tool on critical zones. KeyShot or any rendering pass over the mesh will show you where the geometry looks suspiciously thin — areas that appear translucent or nearly invisible in render are probably too thin for structural use.

Common Pitfalls and What Beginners Miss

The biggest mistake I see is treating the rough projection as the final geometry. It is not. The projection gives you the general shape, but you need to add structural elements separately. The projected surface alone — even at 6mm — cannot reliably support 80–100kg of distributed body weight without additional reinforcement. Think of the projection as a form-finding tool, not a structural one. Another counter-intuitive point: more infill in the 3D print does not necessarily solve the problem. I tested 100% infill in PETG once and the part was rigid but brittle. It cracked along the Z-axis layer lines under repeated loading. The sweet spot for FDM-printed seating structures is usually 40–60% infill with a gyroid or cubic pattern, combined with at least 4 perimeter shells. The perimeters carry the bending loads while the infill prevents buckling between them. Material choice matters more than people think. PETG is forgiving and has decent impact resistance, but it creeps under long-term static load. If this seat is going to hold someone for hours at a time, PLA is actually worse because it becomes brittle at body temperature over time. ABS or ASA handles the creep better, but requires an enclosure to print without warping. If you do not have an enclosed printer, consider switching to injection-molded polypropylene for a production run — it has better fatigue resistance than either PETG or PLA for this application.

When Rough Projection Is the Wrong Approach

There are scenarios where this method will not work well. If you need a seat that supports more than one person simultaneously, the load distribution becomes unpredictable and the projection geometry alone cannot account for the shifting center of mass between two bodies. In that case, a traditional cantilever or leg-based frame design is safer and easier to engineer. Similarly, if you are projecting onto a ground plane that is uneven — outdoor concrete, dirt, gravel — the projection will include those irregularities and create stress concentrations. Flatten the base plane before projecting, or add a separate mounting interface. The process from scan to printable part typically takes about 45 minutes to an hour for someone comfortable with the software. The first prototype print usually fails or needs adjustment, so budget an additional 2–3 hours for redesign and reprinting. If you skip the body scan and use a generic seat template instead, you save roughly 30 minutes but you lose the personalized fit that makes rough projection worth doing in the first place. One final note on downloads or templates: I have not included a downloadable file because the geometry is entirely dependent on the individual's scan data. Generic templates do not transfer well because the sit bone spacing and thigh length vary significantly between people. If you find a free body scanning app and a CAD tutorial that matches your software, those resources will serve you better than a one-size-fits-all model.