Greenhouse Sensors in Winter: Cold, Condensation and Keeping Readings Honest
Share
- Put temperature and humidity probes at crop height in the centre of the house, not by the heater or at the ridge.
- A second sensor in the coldest corner shows the spread across the house, and the spread is the number that matters in winter.
- Condensation on the sensing element makes a humidity probe read near 100 percent and stay there, so the probe must be able to dry.
- Fit white ventilated radiation shields and mount probes where drips cannot fall on the sensing element.
- Compare fixed sensors with a handheld instrument at night, at crop height, at least monthly through the winter.
Table of contents
It was the coldest week of the year, and the greenhouse controller said everything was fine. The temperature sensor by the heater read 18 °C, the humidity looked normal, and the vents stayed shut. Out in the crop, leaves were wet, grey mould was spreading along the lower stems, and the far corner — ten metres from the heater — was sitting at 9 °C. The sensor had not lied — it had measured the warm, dry air beside the heater while the crop shivered in a cold, wet corner it could not see. Winter does not break greenhouse sensors; it exposes where they were never measuring the right thing.
Winter is the hardest season for greenhouse sensing because everything the sensors measure moves at once: temperature falls and stratifies, humidity rises as cold air holds less moisture, condensation forms on every cold surface — including the sensor itself — and the heating system cycles hard, creating warm and cold pockets that shift through the night. A sensor layout that worked in summer, when the house was open and the air mixed, can go quietly blind in winter, when the vents are shut and the air stops moving. Keeping readings honest in winter is mostly about placement, protection, and condensation.
Why winter readings go wrong
The first problem is stratification. Warm air rises, cold air sinks, and in a closed greenhouse in winter the difference between the ridge and the crop can be several degrees. A temperature sensor mounted at head height, or near the heater, measures the warmest air in the house — not the air around the plants. At night the temperature at the sensor swings with each heater cycle while the crop follows a slower, colder curve. The controller thinks the house is warm; the plants disagree.
The second problem is humidity. Cold air holds less water vapour than warm air, so the same absolute amount of moisture reads as a much higher relative humidity when the temperature drops. A greenhouse at 90 percent relative humidity on a cold night is not necessarily wet — but the surfaces in it are cold, and water condenses on them. Condensation on leaves is what drives disease, and condensation on a sensor is what ruins the reading. A humidity probe with water on its sensing element reads near 100 percent and stays there, long after the air has dried. The sensor is not broken; it is wet, and a wet sensor cannot measure humidity.
The third problem is the heating system itself. Heaters create local hot spots and drive convection currents that carry warm air up one side of the house and cold air down the other. Fans meant to mix the air may be off in winter to save energy, or running on a timer that leaves long still periods. Vents open briefly to dump humidity and let in a blast of cold air that drops the temperature near the vent by several degrees for an hour. Each is normal winter operation — and each makes a single sensor's reading a poor guide to the whole house.
Placement: measuring the crop, not the heater
The rule for winter placement is simple: put the sensor where the crop is, at crop height, in the middle of the house — not by the heater, not by the door, not at the ridge. A sensor at plant height in the centre reads the air the plants experience, and tracks disease risk far better than one in the warmest corner. If the house is large, one sensor is not enough: a second sensor in the coldest corner — the far end from the heater, or the north side — shows the spread across the house, and the spread is the number that matters in winter. A house reading 18 °C at the heater and 9 °C in the far corner has a heating problem, and only a second sensor reveals it.
Height matters as much as position. Temperature falls toward the floor on a cold night, and a sensor mounted two metres up can read several degrees warmer than the crop at half a metre. Mount the probe at canopy height, and move it up as the crop grows. Keep it out of the heater's airflow and out of the sun's path through the glazing — a sensor in a sun patch reads the sun, not the air. A simple radiation shield, a small white ventilated cover over the probe, keeps both sun and dripping condensation off the sensing element, and in winter it earns its keep every day.
Humidity probes need the same care, plus one more: they must be able to dry. A probe mounted where condensation drips onto it from the glazing above, or where it sits in a pocket of still, saturated air, will read 100 percent and stay there. Mount it where air moves — near a circulation fan, or in the path of the heater's airflow — and where water cannot drip onto it. Some probes include a heater in the sensing element that periodically drives off condensation; in a winter greenhouse that feature is worth having, because a probe that cannot dry itself will spend half the winter reading saturation whether the air is or not.
Condensation: the sensor's winter enemy
Condensation forms whenever a surface is colder than the dew point of the air around it, and in a winter greenhouse the glazing, the steelwork, and the sensor housing are all candidates. Warm, moist air from the crop meets a cold surface, the air cools, and its water vapour condenses into liquid. The same physics puts water on leaves, which is where disease starts — so condensation is not only a sensor problem, it is the thing the sensor is supposed to warn about. A sensor covered in condensation stops being able to warn about anything.
Protection starts with the housing. A probe in an open, ventilated shield dries faster than one in a sealed tube, and a white shield reflects solar heating that would otherwise warm the probe above air temperature. Cable entries should point downward with a drip loop, so water running down the cable drips off instead of following the cable into the housing. And the sensor should be mounted with the sensing element pointing down or sideways, never up into the path of drips from the glazing.
The other half of the defence is ventilation. A greenhouse with no air movement in winter develops cold, wet pockets wherever the air sits still, and those pockets are exactly where condensation forms and disease starts. Circulation fans keep the air moving gently through the crop, evening out temperature, drying leaf surfaces, and keeping the sensor's microclimate close to the air it should measure. A fan running through the night costs little and does more for reading quality than any sensor upgrade.
Selection: what to look for in a winter greenhouse sensor
Choose on four things: range, protection, the humidity element's ability to recover from condensation, and the output the controller speaks. The temperature range must cover the coldest night the house will see, with margin — a probe that bottoms out at 0 °C is useless in a house that drops below freezing. The housing should be rated for a wet environment, and the humidity element should recover after condensation rather than drifting permanently — worth asking the supplier, because not all elements survive repeated wetting.
A combined temperature and humidity probe in a single housing is the usual choice for a greenhouse, because the two readings belong together: relative humidity only means something at a known temperature, and the controller needs both to decide when to vent or heat. The output — Modbus RS485 or a 4–20 mA loop — must match the controller that will act on it. And the probe should be easy to reach and remove, because a winter sensor needs checking more often than a summer one, and a probe that needs a ladder to reach will not get checked.
Verification: proving the reading in the cold
Verify a winter reading with a second measurement. A handheld thermometer and hygrometer, carried to the sensor's position at crop height, should agree with the fixed sensor within a degree or two — if they do not, one of them is wrong, and the portable instrument is the referee. Check at night, not midday: night is when stratification and condensation are worst, and a sensor that agrees at noon and disagrees at 2 a.m. has a placement or condensation problem, not a calibration problem.
Check the sensor's own condition monthly through the winter: look for water inside the housing, corrosion on the terminals, and dirt or algae on the sensing element. Clean the element gently with clean water and let it dry fully before trusting the reading again — a humidity element that has been wet needs time to recover, and readings taken while it is still drying will read high. Recalibrate or replace a probe that reads high after drying, or one that disagrees with the handheld instrument by more than its normal tolerance. And log the spread between the warmest and coldest sensors in the house: a spread that grows through the winter is a heating or circulation problem developing, and it shows up in the numbers weeks before it shows up in the crop.
Troubleshooting
| Symptom | Likely cause | What to check |
|---|---|---|
| Sensor reads warm while the crop is cold | Probe near the heater or at the ridge; stratification | Move the probe to crop height in the centre; add a second sensor in the coldest corner |
| Humidity stuck near 100% for hours | Condensation on the sensing element | Check for drips onto the probe; let it dry and compare with a handheld; improve air movement |
| Temperature swings with each heater cycle | Probe in the heater's airflow | Move the probe out of the direct airstream; shield it and check the reading at crop height |
| Readings drift after a cold snap | Water in the housing or connector; frost damage | Inspect the housing, cable entries and terminals for water and corrosion; dry and re-seat |
| Large spread between two sensors | Poor circulation; heater undersized for the house | Check circulation fans and heater capacity; look for cold draughts at doors and vents |
| Sensor reads high on a bright winter day | Solar heating of the probe through the glazing | Fit a white ventilated radiation shield; keep the probe out of direct sun patches |
Checklist: winter-proofing greenhouse sensors
- Move temperature and humidity probes to crop height, in the centre of the house, away from the heater and the door.
- Add a second sensor in the coldest corner and log the spread between the two.
- Fit white ventilated radiation shields and mount probes so drips cannot fall on the sensing element.
- Point cable entries downward with drip loops; check the housing for water monthly.
- Keep circulation fans running through the night to even out temperature and dry leaf surfaces.
- Compare fixed sensors with a handheld instrument at night, at crop height, at least monthly.
- Clean the sensing element gently and let it dry fully before trusting the reading again.
- Recalibrate or replace any probe that still reads high after drying, or disagrees with the handheld beyond tolerance.
Winter does not ask a greenhouse sensor to do anything new — it asks it to do the same job in harder conditions, with still air, cold surfaces, and the heating system fighting the weather. The failures are almost never electronic: they are a probe measuring the wrong air, a sensing element that cannot dry, or a house with one sensor where two were needed. A RS485 temperature and humidity sensor, mounted at crop height, shielded from sun and drips, checked against a handheld instrument on a cold night, and read alongside a second sensor in the coldest corner, keeps telling the truth all winter — and in a greenhouse, the truth is what keeps the crop alive.
Our engineers can help you pick the right model, check compatibility, and get your installation right the first time.
RS485 Temperature & Humidity SensorView product →
Related Blogs
A Greenhouse Monitoring Kit: Which Sensors Go In, and What the Alarm Should Catch
Two Humidity Sensors Disagree by 10%: Calibrating with Salt Solutions and Reference Checks
A Weather Station Kit, Unboxed: What's Inside and the Assembly Order That Saves a Day
The Classroom Feels Stuffy but the CO2 Sensor Says Fine: Ventilation Rates from CO2 Data