When a building’s system starts failing.
The useful first question is not which contractor but what is actually wrong. In a high-rise the answer is usually somewhere a committee cannot see, and the way a fault behaves narrows it down faster than any description of the symptom.
QUESTION ANSWEREDWhy does the same door, camera or handset keep failing, and what should be established before quotations are invited?
When a building's security system starts failing, the useful first question is not which contractor to call but what is actually wrong. Faults in a high-rise usually sit in the shared infrastructure rather than the visible device, so a committee that can describe the pattern gets a faster and cheaper answer than one that reports only the symptom.
Why the visible device is rarely the fault
A resident reports that the intercom handset in unit 12-3 is dead. The handset is the only part of the system anyone can see, so it is the part that gets blamed — and it is the part least likely to be at fault. In a high-rise, an intercom, an access control system or a CCTV installation is mostly cable, mostly hidden, and mostly shared. The device on the wall is the end of a chain that runs through a floor distribution point, up a service riser, into a head end in a plant room, and back to a power supply that may be on a different circuit from everything around it.
This matters commercially, not just technically. A building that replaces the visible device has spent money without touching the cause, and the fault returns. A building that establishes where in the chain the fault sits can price the actual repair — and can tell whether it is looking at a small job or the first sign that a system has reached the end of its life.
The four layers a fault can live in
Almost every fault in a building security system belongs to one of four layers. Working out which one is the whole diagnostic job, and a committee can get most of the way there without any test equipment at all.
| Layer | What it is | How it usually shows |
|---|---|---|
| Device | The camera, handset, reader, door station or keypad someone can point at | One unit affected. Everything around it is fine. |
| Cabling | The run from the device to the floor point, and the riser from the floor to the head end | A whole floor, or a whole vertical stack of units, fails together. |
| Head end | The controller, recorder, exchange or server the devices report to | Everything on that system stops at once, across the whole building. |
| Power | The supply, its backup, and whatever shares the circuit | Intermittent. Comes back on its own. Returns weeks later. |
Notice that the pattern of who is affected identifies the layer before anyone opens a panel. One unit is a device. One stack is a riser. The whole block is the head end. Comes and goes is power. That is not a rule of thumb invented for a website; it is the first thing an engineer establishes on arrival, and a committee that has already established it saves the visit.
What to record before anyone attends
The most expensive attendance is the one that ends with we could not reproduce it. Intermittent faults are common in ageing ELV systems and they are the ones a single visit is least likely to catch. A short written record, kept by whoever takes the complaints, changes the odds substantially.
- Which units, exactly. Numbers, not “several on the high floors”. The unit numbers are the diagnosis.
- Which entrance or which camera. A building with three lobbies has three door stations and often three separate cable runs.
- When it started, and whether anything happened around then — a power interruption, lift works, a renovation, a lightning storm.
- Whether it is constant or intermittent, and if intermittent, at what times of day.
- What has already been done, by whom, and what was replaced.
That list costs nothing and is worth more than a photograph of the handset. It is also the beginning of the documentation a building should be keeping anyway — see takeover and maintenance.
How each system tends to fail
The four layers apply to every ELV system, but each has its own characteristic failures, and knowing them shortens the conversation.
- Intercom. Dead handsets in a vertical line point at the riser. A door station that calls but cannot release the door is usually the lock circuit rather than the intercom. Audio that works one way is almost always a cable pair, not a station. Worked through in detail in your intercom has stopped working.
- Access control. Cards that work at one door and not another is a controller or a reader. Cards that stop working for everyone at once is the controller or its database. A door that stays unlocked is a lock or a power-fail setting, and it is the most urgent fault on this list.
- CCTV. A camera with no image at all is usually power or cable; a camera with a poor image is usually the camera. Footage that cannot be retrieved when needed is a recorder, a disk or a retention setting — and it is discovered at the worst possible moment.
- Perimeter and alarm. Repeated false activations are almost never a faulty detector; they are a detector in the wrong place, or one that has drifted out of alignment.
What a diagnostic attendance should give you
A visit that ends in a verbal “we will send a quotation” has not given the building anything it can use. A committee is entitled to leave a diagnostic attendance holding a written record, and should ask for it in advance so there is no argument afterwards.
- N01What was tested, and howWRITTEN
- N02Which layer the fault sits inIDENTIFIED
- N03What was found in the riser or head endRECORDED
- N04Whether the fault is isolated or systemicSTATED
- N05What can be repaired, and what cannotSEPARATED
- N06What is at risk if nothing is doneSTATED
The last line matters more than it looks. A dead handset is an inconvenience. A door that fails unlocked, or a recorder that has silently stopped retaining footage, is a liability the committee is carrying without knowing it.
When a fault is really an end-of-life signal
Some faults are events and some are symptoms. The distinction is not about the age of the system in years — it is about whether the same class of failure keeps arriving from different parts of the installation, and whether the parts to fix it still exist.
Three repairs in eighteen months to three different sections of the same riser is not three faults; it is one ageing cable installation announcing itself. A controller that can only be repaired with parts pulled from another controller is on borrowed time regardless of how well it is currently working. Working out which case a building is in is the subject of repair or replace, and it should be settled before a committee commits to either.
What to do while a decision is pending
Decisions in strata buildings take time: a survey, quotations, a committee meeting, sometimes a general meeting. Meanwhile the building still has to operate. A few interim measures are usually available and are worth asking about explicitly, because a contractor quoting for a replacement has no particular reason to offer them.
- Isolating a failed section so a fault in one stack does not take the rest of the system down with it.
- Confirming that whatever still works is actually recording, and for how long — a CCTV system that retains three days when the committee believes it retains thirty is a serious exposure.
- Making sure any door that has failed is failing in the safe direction for that door, which is not the same answer for a fire exit and a store room.
- Getting the as-found condition written down now, so that the eventual scope of works is based on a record rather than a memory.
Where this leads
A fault is the beginning of one of three conversations: a repair, a replacement, or a maintenance arrangement so that the next one is somebody's defined responsibility. Occhio Tec works on the Extra Low Voltage layer of Kuala Lumpur buildings, and describes its niche as high-rise residential blocks in Mont Kiara, Bangsar and KLCC, much of that work being service and maintenance. The company was established in 2010.
If a system in your building is misbehaving, the practical next step is a site survey: what is there, what condition it is in, and which of the three conversations you are actually having.