Replacing an air handling unit is straightforward when you can turn the building off.
Very few buildings can. Hospitals treat patients. Airports run flights. Manufacturing lines have output targets. Data centres cannot get warm. Retail units trade. In most of the buildings we work in, “we’ll shut down for a fortnight” is not on the table, and the question becomes how to do the work around an operation that keeps running.
That’s a different discipline from a straightforward installation, and it’s worth understanding what it involves before you plan a project around it.
The constraint that shapes everything
In a live facility, the ventilation cannot simply stop.
Sometimes that’s about comfort, and a few hours without ventilation is tolerable. More often it’s about something harder — pressure regimes that must be maintained, contamination control, process requirements, or heat that has to be removed continuously or equipment fails.
Once you accept that constraint, the project stops being a technical exercise and becomes a logistical one. The unit selection is usually the easy part.
We carried out a live-facility replacement at a major London airport, where the terminal remained fully operational throughout.
The four ways to do it
Phasing. Where there’s more than one unit serving an area, you replace them one at a time and the remaining plant carries the load. It’s the cleanest approach, though it usually means accepting reduced capacity while the work is under way and choosing the right time of year to do it.
Temporary provision. Bring in temporary plant to cover while the permanent unit is replaced. It costs more, and the temporary equipment needs somewhere to sit and a route for its ductwork, but it maintains full service throughout. Frequently the only option where the ventilation is critical rather than merely desirable.
Out-of-hours working. Nights, weekends, shutdown periods. This works well in buildings with genuine downtime — offices, schools, some retail — and not at all in facilities that never close.
Sectional replacement. Rather than swapping the whole unit, replace it section by section, keeping the rest running. Slower and more fiddly, but it keeps the system in service and can be the answer where the plant is large and the space is tight.
Most real projects use two or three of these in combination.
What has to be planned before anyone turns up
A proper survey of the existing installation. Not just the unit, but what it serves, what depends on it, and what happens if it stops. In a live facility the consequences of a mistake are borne by the client’s operation, not just their programme.
A method statement that reflects the actual building. Access routes, working hours, noise restrictions, permit requirements. In healthcare and aviation environments this is often the largest part of the planning.
Coordination with the people who run the building. They know things the drawings don’t — which areas are busiest when, which departments will complain, when the quiet fortnight actually falls. Getting them involved early saves more time than any amount of programming.
A contingency plan. Something will not go as expected. What happens if the replacement unit is delayed, or if a fault is discovered when the old plant comes out? In a live building, “we’ll sort it tomorrow” may not be acceptable.
Commissioning within the constraint. New plant has to be commissioned and balanced, and that has to happen without disrupting the areas already back in service.
Access and delivery
Live buildings are frequently the ones with the worst access, because the plant was installed when the building was built and nothing has been designed to get it out again.
Where a conventional delivery isn’t possible, a flat-pack unit assembled in position is often the answer. It takes longer on site, but it avoids building works and keeps the delivery route to something that can pass through an occupied building.
The old unit needs a route out as well, and in a live facility that route probably runs through areas people are using. That has to be planned rather than discovered.
The sectors where this comes up most
Healthcare. Pressure regimes, infection control, and areas that genuinely cannot be taken out of service. HTM 03-01 sets out what’s required, and we’ve written about healthcare ventilation compliance separately.
Aviation. Terminals operate continuously, security constraints affect access and working hours, and passenger areas can’t be disrupted.
Manufacturing and process. Production targets, and processes with their own environmental requirements. Downtime has a cost that’s easy to quantify and hard to justify.
Data centres. Cooling that cannot be interrupted, with equipment failure as the consequence rather than discomfort.
Retail and leisure. Trading hours, customer comfort, and delivery routes that run through public areas.
What to look for in a contractor
Ask about live-facility experience specifically. Installing an AHU into an empty plant room and replacing one in an operating hospital are related but different jobs.
Ask how they’d phase it before you ask what it costs. If the answer is vague, the price won’t hold.
Ask what happens when something goes wrong — because it will, and the answer tells you a lot.
Ask who’s actually on site. Live-facility work depends heavily on the people doing it exercising judgement in the moment.
If you’re planning a ventilation upgrade in a building that can’t stop, the phasing conversation should happen at the same time as the technical one. Have a look at our projects for the kind of environments we work in, or get in touch to talk through the constraints on yours.
