Hollow Sphere of Basic Bricks That Really Rolls

Oct 30, 2026 Last reply: 11 hours ago 3 Replies
Jump to replies (3)

The detail that floored me is that this big hollow sphere is not glued at all. Every brick is held by the ordinary clutch of standard pieces, and the finished ball still rolls smoothly across carpet and stops without cracking. It even came through several runs down a small plastic slide, plus children pressing loose bricks against the surface and checking it for cracks.



Sam Ioannidis worked the shape out in a spreadsheet, with a formula producing the x, y and z coordinates of every block. A few parameters at the top set the ball's size, and the layer patterns and brick counts follow from there. Assembly started from a circular yellow base and grew outward and upward into a hollow bowl, with red stepping inward near the top to close the dome. Pieces alternate direction from one layer to the next to make the structure strong, and the wall averages about 34mm thick, roughly five single units.



The scale is serious: around 8000 bricks at roughly six units each, estimated at about 24 litres of volume, and a whole month of building. Everything came from four big boxes of old pieces bought years earlier for a play cubby house that later came down. The maker says hunting for enough long thin bricks among thousands of pieces was the hardest part.



Not everything is perfect, and that is part of the charm. A forceful test push snapped off the bottom quarter, and during the repair extra pieces were packed inside for strength. Loose bricks can sometimes be heard moving around as it rolls, and it was never opened back up to fix that. The total count was also only ever estimated, never tallied. Next on the list is breaking it apart for an even bigger rolling cartwheel, and I am keen to see that.



Project by Sam Ioannidis. Watch it on their YouTube channel:

formatting link


Alternating the brick direction layer by layer looks like the smartest call in the whole design. With nothing glued, every joint is a potential weak point, and crossing the bonds that way spreads the load instead of letting one seam split open. It probably costs extra sorting time during assembly, but for a ball this size that trade seems well worth it.

One thing I keep wondering about: when the upper layers step inward to close the dome, how does the maker handle the gap between each stepped ring and the curve underneath? Does the same formula simply output shorter spans per row, or is there a separate calculation for those closing rows?

For the bigger cartwheel, it could be worth rigging heavy curtains plus some proper artificial lights before the next round of photos. Here indirect sunlight was needed, since shiny bricks reflect harshly under artificial lighting, and the maker could only shoot in daytime because mixing day and night shots would cause flicker. Controlled lighting would allow filming at any hour with even brightness.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required