Quick answer
Across every industrial chiller breakdown call-out we attended in the 12 months to September 2026, the most common identifiable cause was refrigerant loss, at 27% of the call-outs where the cause is clear from our records. Electrical, control and sensor faults came second at 23%, followed by flow, pump and strainer problems at 16%, and condenser, fan and high pressure faults at 14%. Outright compressor failure accounted for just over 1%. In other words, most breakdowns are not catastrophic mechanical failures. They are leaks, dirt, airflow and flow problems that build up slowly and give plenty of warning.
Everyone in process cooling has opinions about what breaks most often. Very few publish their numbers.
So we went through our own job records for the 12 months to 28 September 2026: every breakdown call-out, what our engineers found and how quickly we got there. No modelling, no industry averages, no estimates. Just what actually happened across our customers’ plant.
Here is what the data says, including the parts that are less flattering.
The Headline Findings
- Median attendance: 2.4 hours from the job being logged, on the jobs where an appointment time was recorded
- 52% attended within 4 hours, and 79% within 24 hours
- June was the busiest month of the year by a clear margin, with July close behind. August was the quietest
- A small number of sites generate a large share of call-outs. The ten busiest sites accounted for around a third of all the breakdowns we attended
- Outright compressor failure was rare, at just over 1% of identified causes
What Actually Breaks
Of every breakdown call-out we attended, the cause is clearly recorded in the engineer’s notes for just under 60% of them. Here is how those break down.
| Cause | Share of call-outs with a clear cause |
|---|---|
| Refrigerant leak or loss of charge | 27% |
| Electrical, controls or sensors | 23% |
| Flow, pump or strainer | 16% |
| Condenser, fans or high pressure | 14% |
| Expansion valve, drier or low pressure | 10% |
| Water, glycol or tank level | 9% |
| Compressor failure | 1% |
1. Refrigerant loss is the single biggest cause
More than a quarter of the breakdowns where we know the cause came down to refrigerant: a leak, a system short of charge, or a machine that had been quietly losing gas for months.
This matters for two reasons. First, a system losing refrigerant does not usually stop dead. It loses capacity, works harder, runs longer and costs more to run, often for weeks, before it finally trips on low pressure. Second, under the F-Gas Regulations a leak has to be found and repaired, not simply topped up, and leak checking is a legal duty for the operator at intervals based on the refrigerant charge. See our F-Gas compliance guide.
A leak check is a scheduled hour’s work. A leak found at 7am on a Monday when production cannot run is something else entirely.
2. Nearly a quarter are electrical, control or sensor faults
Contactors, fuses, phase failures, controllers, probes and transducers. These are rarely dramatic and often cheap to fix, but they stop production just as effectively as a seized compressor.
They are also the faults most likely to be misdiagnosed. A failing probe can look like a refrigeration problem, and a chiller that trips on high pressure because a fan contactor has failed is not a refrigeration fault at all. That is why our alarm code guides spend as much time on what a code does not mean as on what it does.
3. Flow problems are the most preventable thing on the list
Blocked strainers, airlocked or failing pumps, closed valves, and glycol too thick to move properly in winter. Sixteen per cent of identified breakdowns, and a large share of those needed nothing more exotic than cleaning a strainer.
If you do one thing after reading this, check your inlet strainers. See how to identify and resolve chiller flow issues.
4. Condensers and fans: the dirt problem
Fourteen per cent, and heavily weighted towards summer. A condenser coil packed with dust, cardboard fibre or seed fluff cannot reject heat, so the machine climbs on pressure, sheds capacity and eventually trips.
Which is why June was our busiest month by a distance. The chiller did not suddenly get worse in June. The ambient temperature simply exposed what was already there.
5. Compressor failure is rare
Four outright compressor failures across the whole year, just over 1% of identified causes.
That is the most encouraging number in the whole exercise, and the most important one for anyone weighing up whether maintenance is worth the money. The compressor is by far the most expensive component on a chiller, and it very rarely fails without warning. It fails after months of running short of refrigerant, or with poor oil return, or repeatedly restarting against a fault that kept being reset.
Nearly every expensive failure we see is the end of a chain of cheap, visible problems that nobody acted on.
Being Straight About the Data
Two caveats, because a statistic without them is marketing rather than information.
We could not classify around 40% of call-outs. Many jobs are logged with a description as brief as “Breakdown”, and not every attendance has an engineer’s note attached to the asset record. We have reported percentages against the call-outs where the cause is clear, rather than pretending we know about the rest. We are tightening how faults are recorded so next year’s figures cover more of the work.
The attendance figures cover roughly half of the call-outs, being those with a recorded appointment time. We have not extrapolated from them.
What We Would Do With These Numbers if We Ran a Plant
- Treat refrigerant as a leading indicator. If a machine has needed gas, it has a leak. Find it.
- Clean condensers before summer, not during it. June is not a coincidence.
- Put strainers on the daily or weekly check. It is the cheapest breakdown prevention available.
- Stop resetting repeat alarms. A fault that trips, gets cleared and trips again is a fault with a countdown on it. See what your alarm codes actually mean.
- Look hard at the worst sites. A third of our call-outs came from ten sites. If you are having repeat breakdowns, the machine is usually telling you something structural: it is undersized, badly sited, poorly maintained or at the end of its life.
Method
Figures cover breakdown call-outs logged between 28 September 2025 and 28 September 2026 from our own job management system. Causes were categorised from job descriptions and engineers’ asset notes, with each call-out counted once against its primary cause. Attendance times are measured from the job being logged to the appointment. Percentages by cause are calculated against the call-outs with a clearly recorded cause, not against the total.
We intend to publish this annually.
The Underlying Point
Industrial chillers rarely break suddenly. They deteriorate: losing gas, collecting dirt, running with restricted flow, tripping on alarms that get cleared. The breakdown is the last event in a sequence, not the first.
That is the argument for planned maintenance, and it is not our opinion. It is what a full year of call-outs looks like when you count them.
If you want to talk about maintenance on your own plant, our PPM plans start with a free site survey, or you can call us on 0121 820 8946.
