The Performance Gap: Retro-Commissioning Versus a New Chiller
A 9,000 m² office completed in 2019 was designed to an energy use intensity of 85 kWh/m²/yr; metered performance is 118. The west zone heats and cools simultaneously between 14:00 and 17:00 on most weekdays. Air handling units run 24/7 and night setback was overridden during a 2021 tenant complaint and never restored. About 30% of VAV boxes sit at minimum airflow with reheat active. No commissioning records exist beyond handover. One tenant runs a small lab after hours that is not on any after-hours schedule. The utility budget is 11% over. A vendor proposes a USD 600,000 chiller replacement; a retro-commissioning provider quotes USD 85,000.
1. Set out a 60-day diagnostic plan, ordered by what it costs and what it can reveal.
Read the worked answer
Days 1–10 - desk analysis, near zero cost. Pull 12 months of interval utility data and check the shape: does baseload suggest 24/7 operation? Pull BAS trend logs for AHU status, supply-air temperature, valve and damper positions in the west zone, and VAV airflow versus setpoint. Read the original design intent and any tenant fit-out drawings. Confirm which areas are on which schedule.
Days 11–25 - site verification. Night walk to confirm what is actually running at 02:00. Verify the after-hours lab with the tenant and add it to the schedule properly. Measure supply airflow at a sample of VAV boxes with a balometer against design. Check economizer damper linkage and mixed-air temperature. Verify sensor calibration (a failed sensor can be driving simultaneous heating and cooling).
Days 26–45 - quantify and test. Implement reversible trial measures: restore setback with a documented exception for the lab, repair the west-zone reheat/cooling conflict, reset static pressure and supply-air temperature. Trend for two weeks and compare against a weather-normalised baseline.
Days 46–60 - package the findings into an owner-ready list with measured savings, cost, and persistence risk, and present it before any capital decision is taken.
Rationale: the chiller is being asked to carry loads created by schedules and control faults. Diagnose the load before sizing a plant.
2. Build the case for retro-commissioning rather than the USD 600,000 chiller replacement. Show your arithmetic and the assumptions you would have to verify.
Read the worked answer
Illustrative comparison (9,000 m², electricity at USD 0.14/kWh; all assumptions to be validated before decision):
| Retro-commissioning | Chiller replacement | |
|---|---|---|
| Cost | USD 85,000 | USD 600,000 |
| Assumed saving | 12 kWh/m²/yr | 8 kWh/m²/yr |
| Annual kWh | 108,000 | 72,000 |
| Annual saving | ~USD 15,100 | ~USD 10,100 |
| Simple payback | ~5.6 years | ~59 years |
The chiller's measured saving would also likely be smaller, because the chiller load is inflated by 24/7 operation and simultaneous heating and cooling. Fixing the load reduces the chiller's work without any capital.
What must be verified before committing:
- Weather-normalised baseline from interval data, not a degree-day estimate.
- Whether the chiller is actually inefficient or simply oversized (check part-load performance and staging).
- Whether the simultaneous heating/cooling is a control fault, a sensor fault, or a zoning design problem - each has a different fix and cost.
- Whether the plant is at end of life on condition, not just age; get a condition assessment.
Recommended sequence: fund retro-commissioning now, measure for two seasons, then re-size the chiller question against the reduced load. Replacing a chiller sized for a load that no longer exists wastes capital twice.
3. Savings from this kind of work often decay within 18 months. What do you build into the program to prevent that?
Read the worked answer
Decay is a management failure, not a technical one. Build in:
- A measurement and verification plan using utility-bill analysis (IPMVP Option C) with a documented baseline, weather normalisation and a stated uncertainty. Publish monthly, not annually.
- Change control on the BAS. Any schedule or setpoint override requires a ticket with an owner and an expiry date. Overrides that persist become invisible load. Review the override list monthly.
- Written sequences of operation and a live setpoint schedule held by the FM team, updated after every change - so the next technician inherits the intent, not just the equipment.
- A tenant after-hours process. The lab is legitimate; the fix is scheduling, not suppression. Give tenants a simple request route so they do not reach for the override.
- Re-commissioning on a cycle (typically every 3–5 years, plus after any major fit-out) written into the O&M contract and budget so it is not discretionary.
- Alarm reviews and fault detection on the key points: simultaneous heating and cooling, schedule deviation, economiser faults.
- Report persistence, not just project savings, in the same dashboard as the utility budget - the 11% overrun is the accountability anchor.