3D printed basketball debuts in NBA slam dunk contest

• 1
  What matters if you are specifying a printed part
• 2
  How a 3D printed basketball stores and returns energy
• 3
  Cell geometry, strut size, and durometer: the three dials you actually turn
• 4
  Why printing a full-size lattice ball is hard
• 5
  When a lattice part makes sense and when it does not
• 6
  Lattice additive vs CNC machining: pick by function

A normal basketball is a pressure vessel. A butyl bladder holds roughly 7.5 to 8.5 psi, a wound or laminated carcass carries hoop stress, and a leather or composite cover gives grip. Bounce comes from the air spring inside. Lose 2 psi and the ball feels dead on the rim.

The 3D printed basketball removes the air spring. Instead, the whole volume is filled with a repeating lattice: billions of small cells printed from a flexible photopolymer or thermoplastic elastomer. When the ball hits the floor, struts in the contact zone bend, buckle locally, and store strain energy. They spring back on release. Rebound is a material and geometry property, not a gas law.

That shift has a real consequence for engineers. You can no longer tune the ball with a pump. You tune it with cell size, strut diameter, wall thickness, and the durometer of the resin. A 2 mm cell with 0.4 mm struts behaves nothing like a 4 mm cell with 0.6 mm struts, even in the same polymer.

The lattice also spreads impact over thousands of tiny members instead of one pressurized shell. Local dents do not deflate the ball. A deep scuff changes one region's stiffness, and the rest keeps working. That is the core reason the format survived a dunk contest at all.

Cell geometry decides how the lattice deforms. BCC and octet-truss cells bend and stretch under load, which gives a springy, fairly linear return. Simple cubic or grid cells tend to buckle in planes and can collapse suddenly. For a ball, you want gradual, recoverable buckling, so bending-dominated cells are the safer starting point.

Strut diameter drives stiffness fast. Doubling strut diameter roughly multiplies bending stiffness by a factor of eight in a bending-dominated lattice. In practice, a 0.3 mm change in strut diameter is a large change in feel. Print a small coupon grid of strut sizes before committing to a full sphere.

Durometer sets the floor for how soft the ball can feel. Softer elastomer gives more grip and a longer contact patch, but it also creeps under sustained load and runs hotter in the printer. Harder grades hold shape and last longer in fatigue, but the ball can feel like a rock at low impact speeds.

The three dials interact. A soft elastomer with thin struts may bottom out and hit the hard shell behind it. A stiff elastomer with thick struts bounces well but hurts your fingers. There is no single right answer, only a target rebound height and a target grip.

• 1
  Bending-dominated cellsBCC, octet-truss, and similar topologies give predictable, recoverable springback.
• 2
  Strut size firstChange strut diameter in 0.1 mm steps. Stiffness moves much faster than you expect.
• 3
  Durometer secondPick the softest grade that still passes your fatigue coupon test.
• 4
  Skin thickness lastA thin outer skin adds grip and abrasion resistance without locking the lattice.

Printing a hollow lattice sphere means supporting every overhang inside the volume. Most elastomer processes cannot use soluble supports easily, so the lattice is usually designed to be self-supporting: struts meet at nodes with angles that the printer can handle. That constraint shapes the cell choice more than any textbook optimum does.

Layer adhesion is the weak link. Each strut is a stack of partially bonded layers. Under repeated impact, cracks start at layer boundaries and grow along the strut. Print orientation and chamber temperature matter as much as the resin data sheet. A coupon that survives 500 compression cycles may fail at 5,000.

Cycle time is long. A dense lattice sphere takes hours per part, and the failure rate on tall, thin elastomer builds is not zero. That is why the format showed up first as a concept and a stunt, not as a league ball. Volume manufacturing of a solid rubber ball is still far cheaper per unit.

Post-processing is limited. You cannot easily sand inside the lattice. You can wash, cure, and add a sprayed or dipped skin, but any finishing that closes cells changes rebound. Keep finishing to the outer surface.

Lattice parts win when the job needs distributed compliance, low weight, or a single-piece replacement for an assembly of springs and foam. Vibration isolators, helmet liners, prosthetic sockets, drone landing pads, and gripper pads all fit that description. The lattice replaces several parts with one printed body.

Lattice parts lose when the job needs high stiffness, tight tolerances, or a hard, wear-resistant surface. A lattice is a spring by design. If your part must hold ±0.005 mm under load, or slide against another part for a million cycles, the lattice will move and wear in ways you cannot control.

They also lose on cost at volume. Printing thousands of small lattices is slow and per-part expensive compared with injection molding or die casting. The crossover usually sits in the low hundreds of parts, and it depends heavily on part size and material.

The practical test is simple. Ask whether the function is carried by a distributed elastic response. If yes, print a lattice. If the function is carried by a rigid surface, a precise fit, or a hard edge, machine it or mold it.