Two words that get used interchangeably, and shouldn't
Ask a management committee about facade tiles and you will hear "剝落" and "空鼓" used almost as synonyms. They are not. They sit at different points on the same failure curve, and confusing them leads to two specific mistakes: repairing what is visible while ignoring what is dangerous, and budgeting for a patch when the building needs a programme.
The distinction
Spalling — the tile has lost adhesion completely and has detached, or is detaching. It is visible. You can photograph it, count it, and in the worst case find it on the pavement.
Hollowing — the tile is still in place, still looks perfect, but the bond between it and the substrate has failed across some or all of its area. There is a void. The tile is being held by friction, by its neighbours, and by the grout line.
Hollowing is the state a tile is in immediately before it becomes spalling. Every tile that falls was hollow first — often for years.
Why the detection method differs
This is the practical consequence.
| Spalling | Hollowing | |
|---|---|---|
| Visible to the eye | Yes | No |
| Found by visible-light survey | Yes | No |
| Found by infrared thermography | Sometimes (as a secondary signal) | Yes |
| Found by hammer tapping | Yes | Yes, at the point struck |
| Detectable from ground level | Often | Almost never |
A survey that only looks — whether by eye, binoculars or high-resolution camera — finds spalling and misses hollowing entirely. That is not a criticism of visible-light survey; it is a statement about what light can and cannot show.
Hollowing requires either a physical test at the point of contact (tapping) or a thermal reading of how that patch of wall heats and cools (infrared thermography). Tapping is precise where the hammer lands and covers a sample. Thermography is inferential but covers the whole elevation.
How to order the repairs
The intuitive ordering — worst-looking first — is the wrong one.
The ordering that matters is consequence-weighted:
- Anything above a place people are. Entrances, pedestrian routes, balconies, parking bays, outdoor seating. A grade-C finding above the main entrance outranks a grade-D finding above a blank service yard.
- Anything at a boundary. Tiles above a property line create third-party liability exposure that tiles above your own courtyard do not.
- Zones with clustered hollowing. A cluster indicates a systemic bond failure rather than an isolated defect, and clusters propagate.
- Isolated findings on low-consequence elevations. These can genuinely wait for the next cycle.
This ordering is what the building health score is weighted to reflect — hollowing carries the heaviest weight at 30%, precisely because it is the leading indicator of the failure mode that injures people.
The budgeting consequence
Here is where the distinction pays for itself.
A spalling-only survey produces a repair quantity: these 40 tiles need replacing. It is a defined, closed number, and it is almost always an underestimate of the real work — because the tiles about to fail are not in it.
A survey that includes hollowing produces something different: these 40 tiles have failed, these 300 have debonded, and the debonded area grew 12% since last year. That is a five-year programme with a defensible phasing, and it is the sort of number a committee can take to an owners' meeting without being asked to come back next year with a bigger one.
If you have had a tile come down
The immediate steps are containment — cordon the area beneath, and check whether anything above the same zone is visibly loose. After that, the useful question is not "how do we replace that tile" but "how much of this elevation is on the same path".
That is a survey question, and a facade health check answers it. Book an asset assessment or call +886-2-7733-7678.
