Soil Conductivity Sensors: Calibration Solutions, Temperature Compensation and Drift
Share
- A conductivity reading reflects the electrodes, the soil temperature, the moisture around the probe, and the probe's condition, not the soil alone.
- Calibrate against a fresh standard solution at two points, and never pour used solution back into the bottle.
- Report EC temperature-compensated to 25 C, since conductivity rises about two percent per degree Celsius.
- Electrode fouling is the commonest cause of drift, so pull, clean and re-check the probe in a standard solution whenever readings creep.
- An EC reading without a moisture reading beside it is half a measurement, and EC is most meaningful near field capacity.
Table of contents
The grower had calibrated the soil conductivity sensor in spring and trusted it all season. By autumn the readings had crept upward by a third, and the fertigation schedule — driven by those readings — had quietly cut back the feed. The crop looked hungry, the grower blamed the fertiliser, and the real culprit was a probe coated in mineral scale, its electrodes fouled. It had drifted — the failure mode nobody notices until the crop does.
Soil conductivity — electrical conductivity, or EC — is one of the most useful numbers in growing, and one of the easiest to misread. It tells you how many dissolved salts are in the soil solution, which is the grower's proxy for nutrient strength. But the number on the screen is not a property of the soil alone — it reflects the electrodes, the soil temperature, the moisture around the probe, and the probe's own condition.
What a conductivity sensor actually measures
A conductivity sensor measures how easily an electric current flows through the soil solution between its electrodes. Pure water barely conducts; water with dissolved fertiliser salts conducts well. The sensor applies a small alternating voltage across two or more electrodes buried in the soil and measures the resulting current. More dissolved ions mean more current, converted into a conductivity value in mS/cm or µS/cm.
The alternating current matters. A direct current would polarise the electrodes — ions would pile up on one side, gas would form, and the reading would drift within minutes. Good probes use alternating current and, in better designs, four electrodes: two to drive the current and two to sense the voltage, removing fouling and contact resistance from the measurement. Two-electrode probes are simpler and cheaper, but everything that happens at the electrode surface — scale, corrosion, a bubble — shows up directly in the reading. A two-electrode probe suits a clean, monitored installation; a four-electrode design earns its keep where the probe sits in the ground for months without attention.
What the sensor reports is the conductivity of the soil solution in contact with its electrodes — not the conductivity of the bulk soil, and not the nutrient level of the whole bed. The reading depends on how wet the soil is around the probe, because the current flows through the water films between soil particles. Dry soil reads low even when the salt content is high, simply because there is less water to carry the current. An EC reading without a moisture reading beside it is half a measurement, and EC is most meaningful when the soil is near field capacity.
Calibration solutions: the only honest reference
A conductivity sensor can only be as accurate as its last calibration, and calibration needs a reference the sensor can be checked against. That reference is a standard calibration solution: a prepared salt solution with a known conductivity at a known temperature. The most common standards in horticulture are built around potassium chloride (KCl), and the familiar values — 1413 µS/cm is the standard most growers meet first — are traceable references, not arbitrary numbers. A standard states its value at 25 °C, so calibrate at 25 °C or correct for the difference.
Calibration itself is simple, and that is exactly why it gets skipped. Rinse the probe in clean water, immerse it in the standard solution so the electrodes are fully covered, wait for the reading to settle, and adjust the sensor's zero or span until it matches the standard. Then check a second standard at a different value — a single-point check can hide a slope error, where the sensor reads correctly at one value and wrong everywhere else. Use fresh solution: a bottle left open evaporates and its concentration rises. Never pour used solution back into the bottle.
How often? That depends on the soil and the stakes. In a clean fertigation line, a check every few months may be enough; in mineral-rich soil with hard water, monthly checks are the honest interval. The trigger for recalibration is not the calendar but the evidence: a reading that no longer agrees with a lab sample, a sensor that reads differently after cleaning, or a probe that has been out of the ground and put back.
Temperature compensation: the correction everyone forgets
Conductivity rises with temperature — roughly two percent per degree Celsius for most soil solutions — because warmer water lets ions move faster. A soil at 10 °C and the same soil at 25 °C will give readings that differ by nearly a third, even though nothing about the salt content changed. Without compensation, a morning and an afternoon reading of the same bed tell two different stories.
Most modern probes measure temperature alongside conductivity and apply the compensation internally, reporting the value corrected to 25 °C. The compensation is only as good as the temperature measurement, and only as good as the coefficient it assumes. The standard coefficient of about two percent per degree is a good average for typical fertiliser solutions, but the true coefficient varies with the ions present — a nitrate-rich solution and a calcium-rich solution do not behave identically.
The practical checks are simple. Confirm the reported EC is temperature-compensated to 25 °C rather than raw. Compare a compensated reading taken in cold soil with one taken in warm soil from the same bed: they should agree within the sensor's normal variation. If they do not, either the temperature measurement is wrong — a probe half out of the ground measures air temperature — or the compensation is not applied at all.
Drift: why readings wander and how to catch it
Drift is the slow change in a sensor's output when nothing in the soil has changed, and it has a handful of usual causes. Electrode fouling is the commonest: mineral scale, biofilm, and iron deposits coat the electrodes, changing their effective area and the reading with it. Polarisation in a poorly designed probe adds a slow creep. A probe shifted by frost heave or cultivation, or left with an air gap beside it, measures a pocket that no longer represents the bed. Genuine soil change also masquerades as drift: a salt front moving down after heavy irrigation looks like a sensor problem until a lab sample says otherwise.
The way to catch drift is comparison. Keep a reference: a second probe in the same bed, a lab EC test on a soil sample taken beside the probe, or a known standard solution checked against the probe after it is pulled from the ground. Clean the probe when it is pulled — soft brush and clean water, never abrasive — and check it in a standard solution before it goes back in. If cleaning restores the reading, the cause was fouling; if not, the probe is ageing and needs replacing.
Installation is the best drift prevention there is. Bury the probe at the right depth for the root zone, in firm contact with the soil — no air gaps, no loose backfill — and in a spot that represents the bed rather than the wettest or driest corner. Keep it away from fertiliser injectors and drippers, where concentrated solution passes the electrodes and the reading spikes with every irrigation cycle. Mark its position: a probe that cannot be found cannot be checked.
Selection: matching the probe to the job
Choose on four things: electrode design, range, temperature compensation, and the output the rest of the system speaks. Four-electrode probes cost more and drift less; two-electrode probes are cheaper and need more frequent checking. The range must cover the soil you grow in — a probe that tops out below the EC of a heavily fertigated bed saturates and tells you nothing. Temperature compensation should be built in and reported at 25 °C, with the raw temperature available so the compensation can be checked. And the output — Modbus RS485 or a 4–20 mA loop — must match the logger or controller that reads it.
Do not buy on accuracy claims alone: a datasheet figure from a clean standard solution says little about a probe after six months in mineral-rich soil. Ask about the electrode material, whether the probe can be cleaned without damage, and what happens to the reading when the soil dries out.
Verifying a conductivity reading
Trust a reading after three checks. First, the probe check: pull the probe, rinse it, and read it in a fresh standard solution. It should match within tolerance; if not, clean it and check again, then recalibrate. Second, the soil check: take a sample from beside the probe and have its EC measured — a saturated paste extract is the reference method — and the two numbers should agree. Third, the trend check: real soil EC moves slowly and follows irrigation and feeding. A reading that jumps with each irrigation points at placement too close to a dripper; one that climbs steadily for months points at salt accumulation or fouling.
Recheck monthly where the water is hard or feeding is heavy, and always after the probe has been disturbed, a fertiliser change, or a new crop. Write the calibration date on the probe's record.
Troubleshooting
| Symptom | Likely cause | What to check |
|---|---|---|
| Readings creep upward over weeks | Electrode fouling or salt accumulation in the soil | Pull, clean and check the probe in a standard solution |
| Readings jump with each irrigation | Probe too close to a dripper or injector | Move the probe from the wetting front; check readings between cycles |
| Morning and afternoon readings disagree | Temperature compensation missing or wrong | Confirm compensated output at 25 °C; check the temperature reading |
| Reading drops sharply in dry weather | Soil too dry around the probe; air gap at the electrodes | Check soil moisture beside the probe; re-seat the probe in firm contact with moist soil |
| Single-point calibration passes, field readings wrong | Slope error; sensor linear only at one value | Calibrate at two standards; replace the probe if the slope will not hold |
| Reading frozen or erratic after frost | Probe disturbed by frost heave; water in the cable or connector | Re-seat the probe; inspect the cable and connector for moisture |
Checklist: keeping a soil conductivity sensor honest
- Calibrate against fresh standard solution before first use, at two points.
- Confirm the sensor reports temperature-compensated EC at 25 °C, with the raw temperature available.
- Bury the probe in the root zone in firm soil contact, away from drippers and fertiliser injectors, and mark its position.
- Check the probe in a standard solution on a schedule the soil earns — monthly where water is hard or feeding is heavy.
- Pull, clean and re-check the probe after any disturbance, fertiliser change, or new crop.
- Compare readings with a lab soil sample whenever the trend looks wrong.
- Log the calibration date with the probe; a sensor with no calibration history cannot be defended.
- Replace the probe when cleaning no longer restores the standard reading.
Conductivity is a simple measurement that goes wrong in slow, quiet ways: a fouled electrode, an uncompensated temperature swing, a probe sitting in an air pocket. The discipline is equally simple — calibrate against a real standard, compensate for temperature, and check the reading against the soil itself. A RS485 soil conductivity sensor, calibrated on a schedule, installed in firm contact with the root zone, and checked against a standard whenever the trend looks wrong, gives a grower a number worth steering a crop by.
Our engineers can help you pick the right model, check compatibility, and get your installation right the first time.
RS485 Soil Conductivity SensorView product →
Related Blogs
A Soil Monitoring Kit for One Field: Probe, Gateway and Power in a Single Deployment
Nutrient Solution EC Keeps Climbing and Plants Burn: Monitoring EC in Hydroponics
A Greenhouse Monitoring Kit: Which Sensors Go In, and What the Alarm Should Catch
Your Rain Gauge Under-Reads After a Storm: The Drip Test and Adjusting Screws