The BMS service visit that pays for itself - a seasonal approach
For systems integrators and controls engineers who look after existing buildings.
A BMS is commissioned once, but the building around it keeps changing. Tenants move, floors are re-planned, plant is replaced, and every hot or cold spell leaves a few overrides behind. Within a year or two, many systems are running a strategy that no longer fits the building, and nobody is paid to notice.
That gap is the opportunity. A service visit that only checks the panel lights and backs up the database is easy for a client to cut. A visit that finds waste, fixes it and shows the result is much harder to give up, and it creates follow-on work.
Why the traditional visit undersells you
The classic quarterly visit is a walk-round: check the head end, look at a few alarms, take a backup, sign the sheet. It meets the contract, but it leaves the client with nothing they can see. When budgets tighten, invisible work goes first.
Published studies of commissioning existing buildings have found worthwhile savings when drift is corrected. The largest, by Lawrence Berkeley National Laboratory (Crowe, Mills and others, 2020), covers nearly 1,500 North American buildings and found a median whole-building saving of about 6% for existing-building commissioning, with a median payback of 1.7 years - and a range from close to zero to over 50%. Monitoring-based programmes in the same dataset had a median of 9%, and work done outside utility programmes 14%. Almost all the measures were schedules, setpoints, resets, sequences and general control fixes - exactly what an integrator does. Results vary widely from building to building, but the message is consistent: most existing systems have recoverable waste, and finding it is skilled work.
Plan the year around the seasons
Instead of four identical visits, plan four that each have a job:
- Late winter - heating under load. Check space temperatures against setpoints, flow and return temperatures, boiler or heat pump sequencing, weather compensation and optimum start performance, while the plant is working hardest.
- Spring - changeover to cooling. Prove cooling plant, free cooling and economiser logic, chilled water resets, lockouts and the deadband between heating and cooling before the first warm week.
- Summer - cooling under load. Look for low chilled water delta-T, overheating zones, simultaneous heating and cooling, and fans or pumps running flat out.
- Autumn - readiness for heating. Check frost protection, heat source sequencing, compensation curves, pump exercise routines and heating lockouts, and clear the summer's overrides.
The client gets a visible rhythm, and you get predictable, well-timed visits that test the systems when they matter.
What every visit should cover
Whatever the season, a good visit covers the same core:
- Head end and network: software and firmware versions, a tested backup, licences, time synchronisation, user accounts and remote access.
- Controllers and comms: outstation health, communication errors, batteries and panel condition.
- Field devices: a sample of critical sensors checked against a calibrated reference, plus actuators stroked and position feedback confirmed.
- Points in hand: every override listed, with a reason and an owner. Release what no longer needs to be held.
- Schedules and calendars: matched to how the building is used today, including bank holidays.
- Setpoints and limits: compared with the control strategy and any agreed changes.
- Alarms: standing and nuisance alarms reviewed, priorities checked, and routing to the right people confirmed.
- Trends: a look back since the last visit for the classic fault signatures.
Align the content and frequencies with your client's maintenance standard. Many FM contracts specify SFG20 schedules, and criticality will change what is needed.
Let the data choose the work
The most useful hour of a visit is often spent before you arrive. A week of trends shows plant running out of hours, valves passing, loops hunting, sensors flatlined and heating and cooling fighting each other.
Where the client will pay for remote monitoring, rule-based fault detection can do this continuously. The US Department of Energy's Smart Energy Analytics Campaign reported median savings of around 9% for North American organisations using fault detection and diagnostics, with the savings coming from acting on what it found. The model is simple: monitoring finds it, and the visit fixes it.
The report is the product
Clients renew what they can see. End every visit with:
- A one-page summary: what we found, what we fixed, what is left.
- A findings register: each item with a priority, an estimated saving where one applies, and a rough cost.
- Before and after trends for anything you changed.
- Next season's actions.
Keep the language plain. A facilities manager should be able to forward the summary to their finance director without editing it.
Turn findings into projects
Some findings are fixed on the day. Others need budget: a failed sensor batch, a pump that should be on a variable speed drive, missing sub-metering, or a controller that has reached the end of support. These become costed proposals with an estimated payback.
Standards help here. BS EN ISO 52120-1 sets out efficiency classes for building automation and control functions, from D up to A. Saying "your building is broadly at class C - here is what moving to B would take" gives the client a framework rather than a sales pitch. Prove the result afterwards with degree-day-normalised metering and a CUSUM chart.
Package it
A common, workable shape is three tiers:
- Essential: planned visits, backups and a visit report.
- Enhanced: adds remote monitoring, alarm handling in working hours and seasonal recommissioning.
- Optimise: adds fault detection, a monthly energy and fault report, a quarterly review, agreed optimisation hours and a tracked savings target.
Agree response times by priority, what remote first response covers, and when reports are due.
Where to start
Pick one client and one building. Pull a week of trends, run through the checklist above, and write the one-page summary. The first report usually sells the second visit.
Further reading: Seasonal changeover covers the spring and autumn checks in detail, The BMS health check is the natural first visit to a new client, Fault detection on a BMS sets out the classic trend rules, Alarm rationalisation deals with standing alarms, Proving savings covers baselines, degree days and CUSUM, and Selling BMS service covers tiers, SLAs and the annual review. For the foundations, see Commissioning a BMS and Where BMS tuning saves energy.
Related tools and training: the CUSUM savings tracker, degree-day normalisation, energy savings estimator, out-of-hours running cost, VSD savings and project payback calculators are all in tecTools, and the BMS fundamentals, Commissioning good practice and Energy and building control classes courses are in tecTraining.
Published guidance mentioned: SFG20 maintenance schedules (BESA); BSRIA BG 4/2003 BMS Maintenance Guide; BS EN ISO 52120-1:2022; LBNL existing-building commissioning study (Crowe, Mills et al. 2020); US DOE Smart Energy Analytics Campaign reports (2019-2020). Results from published studies vary by building and are not a guarantee of savings.