A Greenhouse Monitoring Kit: Which Sensors Go In, and What the Alarm Should Catch

The grower had decided to monitor the greenhouse properly. The budget was approved, the supplier's catalogue was open, and the list kept growing: a temperature sensor, a humidity sensor, a CO2 sensor, a soil moisture probe, a light sensor, a leak sensor — and then the questions started. Which of these actually needed to be in the first kit, and which could wait? Where should each one go? And what should the alarm do when something went wrong — send a text at 2 a.m., or wait until morning? The grower did not need more sensors. He needed a definition of enough.

A greenhouse monitoring kit is a set of decisions, not a shopping list. The right kit covers the measurements that drive real actions — ventilation, irrigation, heating, and CO2 — with an alarm that catches genuine problems without crying wolf. This article sets out what belongs in a first kit, how the pieces fit together around an all-in-one sensor, what the alarm should and should not catch, and how to install and verify the system so the first season's data is worth acting on.

What Goes In: The Measurements That Drive Decisions

Every measurement in a greenhouse kit should answer a question someone will act on. Temperature and humidity are the foundation, because they drive ventilation and heating, and because their combination determines the disease risk: warm, humid, still air is where grey mould and other fungal problems start. A sensor that reports both from the same probe is the right starting point — two separate probes at different heights will disagree, and the argument about which one is right wastes the season.

Carbon dioxide is the second measurement, and it matters for a different reason. In a closed greenhouse, plants draw CO2 down during the day, and the concentration can fall well below the fresh-air level, limiting photosynthesis even when light and temperature are ideal. A CO2 sensor tells you whether enrichment is working and whether the vents are open too far. It also catches a different failure: a CO2 sensor reading near the fresh-air background in a sealed house usually means the sensor has drifted, not that the plants have stopped breathing.

Soil moisture is the third measurement, and it is the one that saves the most water. A soil probe at root depth tells you when to irrigate and, just as importantly, when not to — and a probe that also reports temperature and conductivity lets you watch the root zone as a whole rather than guessing from the surface. One well-placed probe in a representative bed is worth more than three probes in odd corners, because irrigation decisions are made for a zone, not for a single spot.

Light is the fourth measurement, and the one most often skipped. An illuminance or PAR sensor tells you whether the crop is getting enough light, whether the shade screen is doing its job, and whether the supplemental lights are reaching the canopy. It also catches a failure no other sensor sees: a light sensor that reads near zero at midday means a screen stuck shut or a sensor covered in grime, either of which costs yield every day it goes unnoticed.

An all-in-one shutter-box sensor covers the first, second and fourth of these in a single housing — temperature, humidity, CO2, air quality, noise and light — mounted in one place with one cable and one address on the bus. It is not a replacement for a soil probe, which has to be in the ground, but it is the natural core of a greenhouse kit, because it turns four separate mounting, wiring and addressing jobs into one. The trade-off is honest: a single box measures the air at one point, so it cannot see the far corner of a large house, and if the box fails, several measurements go dark at once. For a small or medium greenhouse, one box in a representative spot is the right core; for a large house, add a second box or a separate sensor at the far end.

What the Alarm Should Catch — and What It Should Ignore

An alarm that fires too often is worse than no alarm, because people learn to ignore it. The design rule is simple: alarm on conditions that need a person, and log everything else. A temperature above the crop's safe limit needs a person. A humidity reading that drifts slowly upward over a week does not need a 2 a.m. text; it needs a note on the morning checklist.

Thresholds need two numbers, not one: a level and a duration. A temperature that spikes above the limit for two minutes when a door opens is not an emergency; a temperature that stays above the limit for thirty minutes is. Set a high threshold with a short delay for genuine emergencies — heating failure, ventilation failure, a stuck vent — and a lower threshold with a long delay for trends that deserve attention the next morning. The same applies to CO2: a brief dip when the vents open is normal, but a CO2 reading that stays near the fresh-air level in a sealed, enriched house means the enrichment has failed.

Hysteresis is the other half of a trustworthy alarm. A threshold without hysteresis will chatter — firing, clearing, and firing again as the reading hovers on the line — and each firing is another notification that teaches the grower to ignore the system. Set the alarm to trigger at one level and clear at a level a few points back, so a single crossing produces a single alert. And grade the notifications: a text or call for the failures that damage a crop within hours, an email or dashboard flag for the trends that matter over days.

Choosing the Kit: One Box or Many Sensors?

The choice between an all-in-one box and separate sensors comes down to the size of the house and the cost of a failure. Separate sensors give you redundancy: if the CO2 sensor fails, the temperature sensor keeps working. They also let you place each sensor where its measurement matters. But each separate sensor is a mounting job, a cable run, and an address to manage, and the total cost of installation often exceeds the cost of the sensors themselves.

An all-in-one box is the better first kit for a small or medium greenhouse, because it arrives as one installed unit with one cable, and it covers the air measurements that drive the daily decisions. Add a soil probe as the second purchase, because soil is the one measurement the box cannot make. For a large house, the honest answer is both: a box as the core, plus separate sensors at the far end and in any zone that behaves differently. And whichever route you take, buy sensors that speak the same protocol, so the whole kit lands on one bus and one dashboard instead of three.

Installation: Where the Kit Actually Goes

The most common installation mistake is mounting the shutter box where it is convenient rather than where it is representative. A box mounted beside the door reads the draught, not the crop. A box above a heater reads the heater. Mount the box at canopy height, in the middle of a representative bay, away from doors, heaters, vents and misting lines, with free air circulation around it. The shutter box's louvres shade the sensors inside while letting air flow, but a box in direct sun or in a dead-air corner will still read wrong.

The soil probe is the second installation decision, and it matters more than the box. Bury the probe at the depth the roots actually use, in a representative bed, with the probe in firm contact with the soil — an air gap around the probe makes it read dry in wet soil. Keep the cable run short and protected, with a drip loop below the connector so water runs off instead of into it, and mark the probe's location so nobody spears it with a fork during the season.

Power and communications are the third decision. A kit on RS485 needs a single bus run with proper termination, and the gateway needs to be where it can reach the network — a gateway in a metal shed with no signal is a common and expensive mistake. If the kit runs on PoE, check the power budget before adding a second box, and label every sensor address, cable run and alarm threshold, so the next person to touch the system does not have to reverse-engineer it.

Verification: Proving the Kit in the First Week

A new kit should be verified in its first week, while the installer is still available and the season has not started in earnest. The checks are simple. First, sanity-check every reading against reality: the temperature against a handheld thermometer, the CO2 against the fresh-air background when the vents are open, the soil moisture against a handful of soil from the same bed.

Second, test the alarm on purpose. Force a threshold — warm the sensor, cover the light sensor, or set a temporary low threshold — and confirm that the notification arrives, that it says which sensor and which reading triggered it, and that it clears when the condition clears. An alarm that has never been tested is a hope, not a system. Third, log a full week and look at the data: check for gaps, frozen readings, and step changes, and check that the daily patterns make sense — temperature rising with the sun, CO2 falling during the day, soil moisture dropping between irrigations.

Troubleshooting: When the Kit Misleads

Symptom	Likely cause	What to check
Temperature and humidity disagree with a handheld reading	Box mounted near a draught, heater or sun patch	Remount at canopy height in a representative bay; check the louvres are clean
CO2 reads near fresh-air level in a sealed, enriched house	Sensor drift, or enrichment failure	Check the enrichment system first; compare the sensor against fresh air
Soil probe reads dry in visibly wet soil	Air gap around the probe, or wrong depth	Reinstall with firm soil contact at root depth; check the cable and connector
Alarm fires, clears, and fires again within minutes	No hysteresis on the threshold	Add a clear level a few points back from the trigger level
Several readings go dark at once	Shared power or bus fault	Check the power supply and bus termination; a single box failing takes its measurements with it

Checklist: A Greenhouse Kit That Earns Its Keep

• Start with temperature and humidity, CO2, soil moisture and light — each tied to a decision someone will make.
• Use an all-in-one shutter-box sensor as the core, and add a soil probe for the root zone.
• Mount the box at canopy height in a representative bay, away from doors, heaters, vents and misting lines.
• Bury the soil probe at root depth with firm soil contact, and mark its location.
• Set alarms with a level and a duration, hysteresis to stop chattering, and graded notifications.
• Verify every reading against reality in the first week, test the alarm on purpose, and log a full week before trusting the data.

A greenhouse monitoring kit is not a collection of sensors; it is a set of decisions made in advance — which measurements matter, where they are taken, and what happens when they go wrong. Get those decisions right, and a single box plus a soil probe will tell a grower more than a dozen sensors mounted in the wrong places. The RS485 IoT shutter box weather sensor with CO2, air quality, temperature, humidity, noise, and light sensors is the core that kit is built around: six measurements in one louvred housing, on one RS485 bus, mounted once at canopy height — and when the alarm goes off at 2 a.m., it will be because something actually needs a person.