Scenarios · Retrofit
Retrofit Building Automation: Planning Around What Already Exists
A practical approach to retrofitting building automation into occupied buildings with existing plant, unknown wiring and incomplete documentation.
Quick answer
Retrofit building automation upgrades control in an existing building without replacing every system at once. The work starts with a survey of plant, wiring, protocols and panel condition, then uses gateways, staged cutovers and local control to keep the building operating during the transition.
Retrofit is an interface problem
In a new building, the automation design can assume clean drawings, spare containment and coordinated trades. In a retrofit, the building is already answering back: plant has been replaced piecemeal, panels have been modified, drawings are incomplete, and the space is occupied. The technical challenge is therefore less about choosing a controller and more about discovering what exists and defining a safe interface to it. A successful retrofit respects the existing plant's own controls, uses the interfaces the vendors actually support, and keeps a working fallback until the new sequence has been proven. The design should be written around constraints, not around a preferred product architecture.
Survey before you scope
The survey should produce a system inventory, a protocol scan, a point list and a condition report. Walk the plant rooms, photograph nameplates and panel internals, record firmware and software versions, and test whether the documented protocol is actually enabled. Check spare capacity in existing panels and containment, and identify circuits that cannot be isolated without affecting tenants. Review the as-built drawings against what is installed. The output is a gap list: what can be integrated, what needs a gateway, what must be replaced and what should be left alone in the first phase. That list is the basis for a credible scope and a realistic programme.
Stage the migration and protect local control
Retrofits rarely succeed as a single cutover. Divide the work into phases that each deliver a useful operating result: metering and monitoring first, then lighting or HVAC sequences, then cross-system interlocks. Keep the existing controllers in place where they are safe and serviceable, and add the new layer alongside them. Run old and new in parallel long enough to compare trends and catch wiring errors. Every phase needs a rollback position, a planned outage window and a named person who can stop the work if the building's operation is at risk. Where a controller cannot be kept, replace it in a planned window rather than under emergency pressure. Phasing also spreads cost and lets the operator learn the new system before the next layer arrives.
Plan for disruption, documentation and spares
The commercial risks in a retrofit are often operational: access to plant rooms, noise, shutdowns, tenant communication and the cost of retesting after a failed cutover. Build those into the programme rather than treating them as site surprises. Label every new cable and point, update the as-built drawings, and leave a spare list that reflects what is actually installed, including obsolete parts that are hard to source. The owner should receive a migration record showing what was integrated, what remains on the old system and what the next phase depends on. Spare parts should be ordered before the first phase, not after the first failure. That record is what prevents a retrofit from becoming a permanent puzzle for the next engineer.
Retrofit constraints and design responses
7 rows| Constraint | Survey question | Design response |
|---|---|---|
| Unknown wiring | Which cables are installed, labelled and tested? | Verify before reuse; plan new containment where doubtful |
| Occupied building | When can each area be shut down safely? | Phase work by zone and define outage windows |
| Mixed vendor plant | Which protocols are enabled and licensed? | Use supported gateways; document each interface owner |
| Incomplete documentation | What do the as-builts omit? | Produce a new point list and panel schedule from the site survey |
| Limited panel space | Is there room for controllers and termination? | Use distributed I/O or a new panel with a clean network link |
| Tenant metering | Who pays for which supply? | Confirm meter points and allocation rules before design |
| Phased budget | Which outcomes must the first phase deliver? | Stage monitoring, sequences and interlocks in value order |
Project images
3 images
Frequently asked
01Do we have to replace the existing BMS?
No. A retrofit can integrate the existing system through a gateway or a supported interface, and replace it later if the condition or licence model justifies it.
02How long does a retrofit take?
It depends on the building, access and phasing. The survey and gap list define the phases; the programme should be built around shutdown windows rather than a single completion date.
03Can we keep the building occupied during the work?
Yes, with zoning, temporary controls and planned outages. Occupied work usually costs more because access, noise and retesting have to be managed.
04What if the drawings are wrong?
Assume they are incomplete until verified. The survey should produce a new point list and panel schedule, and the design should carry a contingency for discovered conditions.
05How do we know the retrofit is complete?
Each phase should end with tested points, demonstrated sequences, updated documentation and a signed acceptance record. Completion is evidence, not a date.
Next step
Bring us the building.
Tell us the building type, systems in scope and the outcome you need. We will map the shortest path to a working system.