Why does a terrazzo floor crack?
The figures, and where each one comes from
| Where the movement comes from | What the floor can absorb |
|---|---|
| A cracked slab | ⛔ nothing on a cement system; the crack telegraphs through |
| Shrinkage in the new slab | joint layout, designed in advance |
| Thermal movement | divider strips on the joint plan, not on a drawing |
| A flexible epoxy topping | absorbs some horizontal movement, and takes a crack membrane |
| An antistatic floor | its conductive mesh is not a movement joint |
Why the processes are not interchangeable
The mechanism is worth stating plainly because it is counter-intuitive. A terrazzo floor is strong and hard and it still follows a crack in the concrete underneath, because grinding bonds the surface layer to the slab so thoroughly that they behave as one piece — which is exactly why the flatness works. The remedy is to put the movement where the floor is meant to move, and to place the joints from the slab joint plan rather than from a drawing made in the office, so the divider lines fall where the movement is. Epoxy is the exception because it is a distinct layer rather than a ground continuation, and that is also why no crack membrane is compatible with a cement system.
What these figures are, and what they are not
That cracks in an existing floor telegraph through monolithic and bonded cement terrazzo, and that no crack membrane is compatible with cement systems, are both from NTMA guidance. ⛔ NO SPACING FIGURE IS GIVEN. GB 50209-2010 requires movement joints to follow the design and requires division joints on large areas, but sets no number; the 6 metre figure in circulation could not be traced to a code and is not reproduced. ⛔ That figure for an antistatic floor's conductive mesh is a different thing and is not a joint.
If you are specifying this
These are the forms we actually supply for this question, with production photographs where we have them: