"How long should a dehumidifier run?" is the wrong question, because runtime is not something you set. It is something the equipment reports back to you about your building.
Read correctly, it is the cheapest diagnostic available: it tells you whether the machine is correctly sized, whether the load matches what was assumed, and whether capacity has degraded — using data most controllers already record.
Runtime Is an Output
For a machine cycling on and off to hold a setpoint, the relationship is simple:
Runtime fraction ≈ moisture load rate ÷ capacity at the operating conditionA space generating 100 litres per day, served by equipment delivering 150 litres per day at that space's actual condition, will run about 67 % of the time. The same equipment in a space generating 140 litres per day runs about 93 %. Nothing about the machine changed.
Two points make this useful rather than merely arithmetic.
Capacity must be the capacity at the operating condition, not the catalogue figure. Rating conditions differ substantially between standards and are usually more favourable than real operating conditions — the subject of what a daily water removal rating really means. Using a catalogue number here will produce a runtime estimate that is far too optimistic.
Runtime therefore measures load. Since capacity is roughly known, observed runtime is a direct read on how much moisture the space is actually generating. That is information no other simple measurement provides.
What Normal Looks Like
A correctly sized installation follows a recognisable pattern.
Pull-down. On start-up, or after the space has been open, the equipment runs continuously until the setpoint is reached. Continuous operation here is correct and expected. How long it takes depends on how far the space is from target and on the load — and if a project has a defined pull-down requirement, that requirement may size the equipment rather than the steady-state load.
Steady state. Once at setpoint, runtime falls to whatever fraction matches the ongoing load. In most commercial applications this settles somewhere well below continuous.
Seasonal variation. Runtime should track the load through the year — higher in humid seasons, higher when ventilation rates rise, higher during wet processes. A flat runtime across seasons in a space whose load clearly varies suggests the equipment is at a limit in one direction or the other.
Daily variation. Occupancy, process schedules, door openings and lighting schedules all move the load within a day. Runtime should show that structure.
The shape of the curve is as informative as the number. Runtime that varies sensibly with known load drivers indicates a system that is tracking its space. Runtime that is flat, or that does not respond to changes that should affect it, indicates something saturating.
Continuous Operation: Two Very Different Cases
Equipment running continuously is not automatically a problem. The distinction is whether it is holding setpoint.
Continuous and holding setpoint. The equipment is correctly sized and operating at its limit. This is acceptable in principle, but it leaves no margin: any increase in load, any capacity degradation, or any change in operating condition will cause the setpoint to be missed. It is worth understanding whether the situation is seasonal or permanent.
Continuous and not holding setpoint. Something is wrong, and there is a short list of causes.
| Cause | How to distinguish |
|---|---|
| Load higher than assumed | Compare measured condensate volume against the design load calculation |
| Capacity lower than assumed | Compare against the commissioning baseline at a similar operating condition |
| Operating condition colder than design | Check entering-air temperature; capacity falls sharply as it drops |
| Ventilation or infiltration increased | Check ventilation rate, door discipline, envelope changes |
| New moisture source | Process change, leak, wet materials introduced, drainage failure |
| Setpoint unachievable in current conditions | Check whether the target dew point is below what the equipment can produce |
| Air distribution failure | Check whether conditions are uniform; local humidity with satisfied sensor indicates distribution |
| Degraded capacity | Filter, coil, refrigerant charge, fan — see dehumidifier maintenance |
Working through this list in order is faster than replacing equipment. The most common findings in practice are an unrecognised ventilation or infiltration load, and a fouled filter or coil.
Short Cycling Is the Worse Fault
Very short runtime is usually more damaging than very long runtime, and it is easy to misread as good news.
Why it happens. Oversized equipment for the load, a control deadband that is too narrow, or a sensor positioned where it responds to the machine's own discharge rather than to the space.
Why it matters. Three effects compound:
- Start-up losses. A refrigerant coil takes time to reach the temperature at which it condenses effectively. Short cycles spend a large fraction of each run cooling without dehumidifying.
- Re-evaporation. When the compressor stops while the fan continues, air passes over a wet coil and some condensate that has not yet drained evaporates back into the space — returning moisture that was already removed.
- Component wear. Frequent compressor starts are harder on the machine than continuous running.
The result is that a short-cycling installation can deliver considerably less moisture removal than its runtime fraction suggests, while appearing to satisfy its setpoint. This is the same mechanism that makes oversized air conditioning produce clammy spaces, described in temperature and humidity control in HVAC.
What to do. Widen the control deadband, move the sensor away from the discharge, or if the equipment is genuinely oversized, use staged capacity so that a smaller increment runs for longer periods rather than a large one running briefly.
Using Runtime as a Diagnostic
The practical value of this is distinguishing between two situations that look identical from the outside: the load has increased, or the equipment has weakened.
The method requires one thing recorded at commissioning: a baseline of capacity, power draw, and entering and leaving air conditions at a known operating condition.
With that baseline:
If runtime has risen but condensate per running hour is unchanged, capacity is intact and the load has increased. Investigate ventilation, infiltration, process changes and new moisture sources.
If condensate per running hour has fallen at a comparable operating condition, capacity has degraded. Investigate filters, coil fouling, airflow, refrigerant charge and fan performance.
If both have changed, treat them separately rather than assuming one explains the other.
Condensate per running hour is the key derived figure, and it requires only a runtime record and a condensate measurement. Where condensate is drained rather than collected, an inline meter or a temporary collection period provides the data.
Without a baseline, none of this is available, which is why the commissioning record matters more than it appears to at the time.
Control Strategy Affects Runtime
Deadband. The gap between the setpoint at which the machine starts and the point at which it stops determines cycle length. A narrow deadband gives tighter control and shorter cycles; a wider one gives longer cycles and better effective moisture removal. Where the application tolerates some variation, widening the deadband usually improves performance.
Continuous fan versus cycling fan. Running the fan continuously improves air mixing and sensor accuracy, but drives the re-evaporation loss described above whenever the compressor is off. Cycling the fan with the compressor avoids that loss but reduces mixing. Which is preferable depends on whether uniformity or maximum removal matters more in that space.
Setpoint choice. A setpoint tighter than the application requires increases runtime and energy for no benefit. Where the requirement is condensation prevention, a dew point setpoint referenced to the coldest surface — as set out in condensation risk and dew point — usually permits a less demanding target than an equivalent relative humidity setpoint chosen for safety.
Scheduling. Where the load has a known daily structure, running against that schedule is more effective than reacting to it. In lit growing environments, for example, the moisture load follows the photoperiod closely.
When Runtime Says to Review the Selection
Runtime indicates a selection problem rather than an operating problem when:
- Continuous operation fails to hold setpoint at conditions the equipment was specified for, and the diagnostic list above has been worked through.
- Runtime is at or near continuous during the mild part of the season, leaving nothing in reserve for the demanding part.
- Short cycling persists after the deadband has been widened and the sensor relocated, indicating genuine oversizing.
- Runtime has increased over successive years while condensate per running hour has remained constant, indicating a load that has grown beyond the original design basis.
- The space cannot be brought from an initial condition to target within the time the operation requires, even though steady-state performance is adequate — a pull-down requirement that the original sizing did not address.
The selection sequence to revisit is in commercial dehumidifier selection.
What to Record
- Runtime, logged as hours per day or as a percentage.
- Condensate volume, either collected or metered.
- Space temperature and relative humidity, logged together at sufficient resolution to see cycles.
- Entering-air condition at the equipment, which is what governs capacity.
- Outdoor conditions, to correlate load with weather.
- Ventilation rate and any changes to it.
- Setpoint and deadband settings, and any changes.
- Filter changes, cleaning and service events, with dates.
- The commissioning baseline, retained and accessible.
Hourly logging over a season is far more useful than detailed logging over a week, because the questions that arise are usually seasonal.
Discussing a Project
Yakeclimate designs and manufactures industrial dehumidification equipment for complex climate applications, with a focus on agriculture and energy projects.
We co-develop application-specific dehumidification equipment around the operating conditions, interfaces, and integration requirements of the wider project or system.
Explore industrial dehumidifiers, review industrial ceiling dehumidifiers, or contact Yakeclimate to review the operating conditions for your project.
FAQ
Frequently Asked Questions
How many hours a day should a dehumidifier run?
There is no correct number, because runtime is determined by the space rather than set by the operator. Runtime fraction is approximately the moisture load rate divided by the equipment's capacity at the actual operating condition. A space generating 100 litres per day served by equipment delivering 150 litres per day at that condition will run about 67 % of the time. What matters is whether the setpoint is held and whether runtime responds sensibly to known load drivers.
Is it bad if the dehumidifier runs continuously?
It depends on whether it is holding setpoint. Continuous operation that holds setpoint means the equipment is correctly sized and operating at its limit — acceptable, but with no margin for load increase or capacity degradation. Continuous operation that fails to hold setpoint indicates a problem: higher load than assumed, colder operating conditions than design, increased ventilation or infiltration, a new moisture source, degraded capacity, or an air distribution failure.
Why does my dehumidifier turn on and off constantly?
Short cycling usually means the equipment is oversized for the load, the control deadband is too narrow, or the sensor is positioned where it reads the machine's own discharge. It is more damaging than long runtime: the coil spends much of each short cycle cooling without dehumidifying, condensate re-evaporates from the wet coil when the compressor stops while the fan runs, and frequent starts wear the compressor. Widening the deadband, relocating the sensor, or using staged capacity all help.
How can I tell whether the load increased or the machine weakened?
Compare condensate collected per running hour against the commissioning baseline at a similar operating condition. If runtime has risen but condensate per running hour is unchanged, capacity is intact and the load has grown — investigate ventilation, infiltration and process changes. If condensate per running hour has fallen, capacity has degraded — investigate filters, coil fouling, airflow and refrigerant charge. This diagnostic requires a commissioning baseline, which is why recording one matters.
Should the fan run continuously?
It is a trade-off. Continuous fan operation improves air mixing and gives more representative sensor readings, but whenever the compressor is off, air passing over the wet coil re-evaporates condensate back into the space. Cycling the fan with the compressor avoids that loss but reduces mixing. Spaces where uniformity is the priority favour continuous operation; spaces where maximum moisture removal is the priority favour cycling.
Does a lower setpoint remove more moisture?
It increases runtime and energy consumption, and it removes more moisture only to the extent that the space genuinely requires a drier condition. Where the underlying requirement is condensation prevention, expressing the target as a dew point referenced to the coldest surface usually permits a less demanding setpoint than a relative humidity figure chosen conservatively — with the same protection and lower runtime.
References
- AHAM DH-1, Dehumidifiers — capacity rating standard (US)
- ASHRAE Handbook — Fundamentals, psychrometrics and moisture load calculation
