Eight ELV integration mistakes that survive into commissioning — and how to catch them earlier

ELV integration failures are the most predictable category of building-systems snag, and they are also the hardest to surface in tender review. They survive into commissioning because the seam-level coordination — the cable that crosses three trades, the relay that two panels both think they own, the matrix that nobody re-tested after a sub-contractor change — is nobody's contractual responsibility under a trade-by-trade procurement. They are caught at design stage only when the integrator is held to single-contract coordination across the disciplines. Below are the eight failure modes we see most often, drawn from our own audit work and from take-overs of inherited systems.

## 1. Missing cause-and-effect documentation

The most common ELV integration failure is the absence of a written cause-and-effect matrix. The fire-alarm panel is programmed; the lift-homing circuit is wired; the access-control doors are configured to release on alarm; the PA voice-evac is set up — but none of it is in a single document that ties the inputs to the outputs. We catch this on every audit by extracting the as-is matrix from the panel programming and walking it through with the operations team. The fix is to write the matrix, sign it, and re-test on every system change. The discipline is in the writing, not the panel.

## 2. Shared-pathway crowding

ELV cabling routinely shares pathways with security cabling, fire-alarm cabling and structured-cabling. On retrofit work where the pathway is fixed, we routinely audit pathways that are physically over-crowded — bend radii violated, separation distances ignored, EMI coupling between the fire-alarm loop and the CCTV camera feed corrupting both. We flag this at design stage by mapping the pathway capacity against the cable count and the EMI separation requirements; the fix is sometimes a second pathway, sometimes a re-routed fire-alarm loop.

## 3. Fail-state confusion at access doors

The fail-safe vs fail-secure decision is per-door (see our separate insights piece on this). The most common failure is a door specified as fail-safe in the access-control configuration but installed with a fail-secure electric strike, or vice versa. The fire-alarm matrix says the door releases; the hardware says it does not. We catch this by reconciling the access-control door schedule against the hardware as-built and the fire-alarm matrix on every commissioning, and we re-check on every quarterly AMC visit.

## 4. CCTV retention storage under-sized

Storage is sized by the formula retention × bitrate × camera count. The most common failure is to size against the procurement-stage bitrate (often 2 Mbps for 4MP cameras) without accounting for the actual encoder behaviour during high-motion scenes (bursts to 6 Mbps in busy lobby cameras). The result is storage that runs out at 18 days when the spec said 30. We size on a per-camera bitrate model with motion-burst headroom, and we measure actual storage consumption at week 4 of operation against the model.

## 5. AHU-damper-vs-OT-anaesthesia conflict (hospital-specific)

On hospital deployments, the fire-alarm matrix has to balance damper-closure for fire-safety against OT-ventilation continuity for surgical-anaesthesia safety. The most common failure is to close all AHU dampers immediately on a fire-alarm trigger, including the OT zones, which compromises the anaesthesia delivery for any surgery in progress. The fix is a documented OT override on the matrix, with a clinical-engineering acknowledgement window before the OT-zone dampers close. We caught this on the Tinsukia hospital design and the matrix carries the override explicitly.

## 6. Voice-alarm priority not hardware-enforced

PA voice-evacuation priority over routine paging must be hardware-enforced, not software-configured (see the PA design article). The most common failure is a software-configured priority that a routine paging operation can over-ride if the operator presses the wrong button. The fix is a hardware priority input on every zone amplifier, wired to the fire-alarm voice-evac module, with the routine PA on a lower-priority input. We test the priority on every commissioning by triggering a fire-alarm event during a routine page and verifying the routine page is interrupted.

## 7. Lift-homing not re-tested after lift maintenance

Lift-homing on fire-alarm trigger is one of the most-cited cause-and-effect outputs and one of the most-frequently-broken. The lift maintenance contractor disables the homing circuit during a service visit (to prevent accidental homing) and forgets to re-enable it. The fire-alarm matrix says the lifts home on alarm; the lifts in service do not. We catch this with a quarterly AMC routine that re-tests every cause-and-effect output, including the lift-homing, and a written test record signed by the lift-maintenance lead.

## 8. CCTV analytics that misfire and get disabled

CCTV analytics that misfire repeatedly are disabled by the operations team, at which point the analytics layer was a wasted spend. The most common failure is to push generic loitering and intrusion analytics across the camera estate without per-camera tuning. The fix is to script the analytics rules per camera, walk them through with the operations team, modify per the meeting, and only then push them to the VMS. The discipline is that the operations team owns the rule-set, not the integrator.

## Callout — what buyers most miss

**The integration discipline is contractual, not technical.** Most ELV integration failures are not engineering errors — they are coordination errors that survive because no single party is accountable for the seams between disciplines. Tender for single-contract integration with a written cause-and-effect deliverable, and the eight failures above are catchable at design stage. Tender for trade-by-trade procurement with no integration owner, and at least three of the eight will be in the snag list at handover.

## References

1. NBC 2016, Volume 2 — fire and life-safety provisions.

2. IS 2189-2008 — *Code of practice for installation of automatic fire detection and alarm system*.

3. IS 16102:2014 — *Voice Alarm Systems — Sound systems for emergency purposes*.

4. NABH 5th Edition Accreditation Standards for Hospitals — clauses on facility management and emergency preparedness.