Where should the temperature and humidity probe go?

The short answer

• At canopy height, level with the top of the plants, because that is the air the leaves live in. A probe at the top of the tent reads the lamp’s air; one on the floor reads the pot’s.
• Out of the lamp’s direct beam. A sensor in the light warms up and reports its own temperature. We measured 2.6 °C too high on our own tent from nothing but moving it into the beam. The shaded side of a tent pole is the right spot.
• Out of a dead corner and out of the fan’s direct blast. Still air over-reads humidity; a fan straight on it under-reads temperature. Somewhere the air moves past it gently.
• Not over the canopy if you can avoid it. Anything you hang above the plants takes light away from them.

Why the beam matters more than anything else

A temperature sensor does not measure the air. It measures itself, and it assumes it is the same temperature as the air around it. Put it under a grow light and that assumption fails: the sensor body absorbs radiation the air does not, warms above the air, and reports the warmer number. That is not a defect in any particular sensor; it is what every unshielded thermometer does in sunlight, and a grow light is a small sun.

On our own tent the error was 2.6 °C — the difference between the same probe on the lit side of a pole and the shaded side, minutes apart, same air. Every VPD number, every “too hot” or “in range” verdict, and every fan or humidifier decision that rests on that reading was wrong by that much. If you have a controller that seems to run the fan when the tent feels fine, this is the first thing to check.

Weather stations solve it with a white, louvred radiation shield and a fan that pulls air across the sensor. In a tent the cheap version is the same idea: the shaded side of a pole, with the room’s own air movement doing the aspirating.

The second error, if your probe is a powered gadget

A wireless sensor has a radio and a processor on the same board as the sensing element, and they make heat. Co-located with a plain reference thermometer — same air, same height, so nothing else could explain it — ours read about 2.2 °C high from its own electronics. We fixed it by putting the board to sleep between readings so the radio is off most of the time, and the offset went away. If your probe is a small board in a box that never sleeps, suspect it: put a dumb alcohol thermometer next to it for an hour and compare.

Why canopy height, and not the top of the tent

The air in a tent is not one temperature. The top, near the lamp, is warmest and driest; the floor is coolest and dampest; and the leaves are in between, in whatever layer the canopy happens to occupy. The number you want is the one the plant experiences, so the probe goes level with the top of the canopy and moves up as the plant grows. A probe zip-tied to the tent frame at eye level in week two is reading the wrong air by week six.

Why not over the plants

The place a sensor “sees” best is straight above the canopy, and that is exactly where it costs the most. Anything hung over the plants subtracts light from the leaves under it, and light is the yield lever. It is also a measurement problem: a leaf under a sensor is a shaded leaf, so if that sensor is reading the leaf’s temperature it is reading a leaf cooler than the rest of the canopy — and under-calling the heat stress it exists to catch. We learned that one the hard way on our own leaf sensor. Mount off to the side, on a pole, and aim inward.

“All this to hit 50 % at 23 °C?”

A question that shows up whenever someone lists their fans, dehumidifier and controller and asks whether it is really that hard. Two things make it harder than it looks, and neither is the equipment.

You cannot vent below the room you vent from. Pulling air from the lung room drives the tent toward the intake air’s absolute humidity — the grams of water in it — not its relative humidity. A cool tent fed from a warmer room lands above the room’s RH; a warm tent fed from a cooler room lands below it. If the room is humid, no amount of exhaust reaches 50 %; the fan just runs. That is psychrometrics, not a cannabis fact, and it is why a dehumidifier in the lung room sometimes does more than one in the tent.

The number you read at noon is not the number the plant lives at. When the lights go off the tent cools and relative humidity climbs with no water added, because cooler air holds less. A tent that reads 50 % under the lamp can sit at 70 % all night. If you only ever read the probe when you are in the room, you are reading the easy half of the day. Log it, or read it once at lights-off and once before lights-on, before you decide the equipment is failing.

How to check this yourself

• The pole test. Read the probe on the lit side of a pole, move it to the shaded side, wait five minutes, read again. The difference is your beam error.
• The reference test. Put a plain thermometer next to the probe, same height, for an hour. A steady offset with nothing else changing is the probe heating itself.
• The height test. Read at the floor, at canopy height and at the top of the tent. If they differ by more than a degree or two, the placement matters more than the sensor.
• The overnight test. Read at lights-off and just before lights-on. If those two numbers surprise you, the noon reading was never telling you much.

What would show that we have this wrong

If the pole test shows no difference, your lamp is not putting much radiant load on the sensor and the beam rule matters less for you than it did for us. And if the lung room is genuinely dry and the tent still will not come down, the moisture is coming from inside — a big transpiring canopy, a wet medium, an open reservoir — and the probe is telling you the truth.

Where our device fits, stated narrowly

Croplock’s climate pod is a probe like any other, and every rule on this page applies to it; the setup card says the same thing in fewer words. What the device adds is the record: it reads at canopy height every minute, so the overnight climb and the lights-on dip are a trace rather than a guess, and if you also put a pod in the lung room it can say plainly when venting has no more to give. It does not know where you mounted it — that part is yours.

Sources

• The 2.6 °C beam error and the ~2.2 °C self-heating offset are our own measurements on our own bench (August and July 2026) and are given as such.
• That an unshielded thermometer over-reads under radiant load, and that a vented space tends toward the intake air’s absolute humidity, are general instrument practice and psychrometrics, not cannabis findings; we have not cited a chapter for them because we have not read one that says it about tents.
• The overnight humidity rise is first-party observation from our own tent.

More of these at Answers. The device is at croplock.com.

Related: Leaf temperature and VPD — why the air reading is only half of it. Bud rot: how humid, for how long? — what the overnight trace is for.

Also read: Clones not rooting, seedling stalling · Curling leaves: heat or hunger?