A hydroponic grower calibrates the pH sensor on Monday. By Thursday, a handheld meter says the reservoir is at pH 6.4 while the inline sensor still reads 5.9. The controller, trusting the inline reading, has been holding back acid for three days. Nobody touched the sensor, and the glass bulb looks clean.
In most cases like this, the glass is not the problem. The fault sits on the other half of the sensor: the reference electrode and the salt bridge that connects it to the solution. How large that salt bridge is, and how much junction capacity sits behind it, has more influence on how long a pH sensor stays accurate in nutrient solution than almost any other design choice.
The short answer is this. In hydroponic nutrient solutions, a sensor with a larger salt bridge and larger junction capacity holds its calibration longer, because it resists clogging from precipitates and biofilm and has more reference electrolyte in reserve. Smaller junctions protect the reference better in clean, low-conductivity water, but they clog quickly in nutrient solution, and a clogged junction is the most common reason an inline pH reading stops matching reality.
How a pH sensor actually measures
A pH sensor is two electrodes working as a pair. The glass electrode develops a voltage that changes with hydrogen ion activity in the sample. The reference electrode is supposed to hold a constant voltage regardless of what the sample is doing. The sensor reports pH from the difference between the two.
For that measurement to work, the reference electrode needs an electrical connection to the sample. That connection is the salt bridge, a porous junction through which ions from the reference electrolyte (typically potassium chloride) make contact with the process fluid. Anything that changes the conditions at that junction, whether a partial blockage, contamination seeping inward, or depleted electrolyte, shifts the reference voltage. The sensor cannot tell the difference between a real pH change and a shifted reference, so the reading moves even though the solution has not.
Small salt bridge versus large salt bridge
The size of the salt bridge and junction sets a trade-off between protection and robustness.
A small salt bridge restricts exchange between the reference electrolyte and the sample. The electrolyte depletes slowly, and contaminants have a harder time diffusing inward to poison the reference. The cost is a high-resistance pathway. In very low-conductivity water such as condensate or RO permeate, that resistance makes readings noisy and slow to settle. A small junction also has very little open area, so oils, suspended solids or precipitates can block it quickly.
A large salt bridge presents a much bigger contact area to the sample. It establishes a stable, low-resistance connection and is far harder to block, because fouling has to cover a much larger surface before it disrupts the junction. The trade-off is that a larger junction exchanges more with the sample, so the design needs a correspondingly larger electrolyte reserve to avoid depleting early, and it offers less of a barrier against aggressive chemicals moving inward.
Neither design is universally better. The right choice depends on what the process fluid will do to the junction over months of continuous immersion.
What nutrient solution does to a pH sensor
Hydroponic nutrient solutions typically run at an EC of 1 to 2 mS/cm. That is a well-conducting solution, comfortably above the range where junction resistance causes noisy readings, so the small-junction advantage in low-conductivity water does not apply here. The threats in a nutrient reservoir are different, and they all act on the junction.
Precipitation. Nutrient solutions carry high concentrations of calcium, phosphate and sulphate. When pH rises or the solution is concentrated, calcium phosphate and calcium sulphate can precipitate, and the junction surface is one of the places they deposit. A small junction can be sealed by a thin layer of scale.
Biofilm and organics. Warm, nutrient-rich recirculating water supports microbial growth. Biofilm forms on every wetted surface, including the sensor, and slowly coats the junction.
Continuous immersion. An inline sensor in a recirculating system never rests. The reference electrolyte is in constant exchange with the sample, so the size of the electrolyte reserve directly determines how long the sensor stays stable.
Against all three, the larger salt bridge wins. It takes far more deposit to block, and a design built around a large junction carries more electrolyte to sustain it. This is why a large-junction sensor holds its calibration noticeably longer in a nutrient reservoir than a small-junction sensor of the same quality.
Why a shifting pH reading costs more than a recalibration
In an automated system, the pH sensor is not just an indicator. It drives the acid or base dosing. When the reference shifts, the controller doses against a false reading with complete confidence.
Most crops in hydroponics are held in a pH band of roughly 5.5 to 6.5, because nutrient availability changes outside it. As pH climbs above the band, iron, manganese and phosphorus become progressively less available, and deficiency symptoms appear even though the nutrients are physically in the solution. A sensor reading 0.4 units low lets the true pH settle outside the band while the controller reports everything is in order. By the time leaf symptoms reveal the problem, the crop has already lost growth.
How the Pyxis ST-710 is designed for this
The Pyxis ST-710 series is built around the large-junction approach. It combines a flat pH bubble, which resists contamination in turbid water better than a protruding bulb, with an enlarged salt bridge and large junction capacity aimed at longer service life with minimal maintenance. It outputs 4-20 mA or RS-485 Modbus directly from the sensor body, so it connects to most controllers and PLCs without a separate transmitter.
Two body options suit different lifecycle preferences.
ST-710 (CPVC body). A sealed design that needs no electrode change and no KCl refill. It is used until the end of its life and then replaced as a complete unit. This suits operators who want the simplest possible maintenance routine, with service life depending on the application and solution chemistry.
ST-710SS (stainless steel body). A 304 stainless steel body with a replaceable electrode head, the EH-710. The electrode head carries 3 to 4 times the electrolyte content of a conventional replaceable electrode. When the electrode reaches the end of its life, only the head is replaced and the stainless body stays in service, which lowers long-term replacement cost on larger installations.
Keeping the sensor accurate: cleaning
The flat bubble reduces how much material accumulates on the glass, but no sensor in a nutrient reservoir stays clean indefinitely. When readings become sluggish or calibration needs a larger adjustment than usual, inspect the sensor first.
Mineral scale and biofilm can usually be removed with a short soak, typically 10 to 15 minutes, in dilute hydrochloric acid or a dedicated pH electrode cleaning solution. Rinse thoroughly with clean water afterwards, avoid scrubbing the glass, and recalibrate with fresh buffers before returning the sensor to service. Always follow the cleaning procedure in the Pyxis user manual for the specific model.
Keeping the sensor accurate: diagnostics
Cleaning on a fixed schedule works, but knowing the sensor’s actual condition works better. The ST-710 series supports wireless diagnosis and calibration through the uPyxis mobile and desktop apps, connected via a Pyxis Bluetooth adapter.
The diagnosis screen gives access to the raw electrode signal in millivolts, and two values are worth tracking over time. The first is the offset, the millivolt reading at pH 7, which should sit close to zero for a healthy electrode. A steadily growing offset points to a contaminated or depleting reference junction. The second is the slope, the change in millivolts per pH unit, which for an ideal electrode is about 59 mV per pH unit at 25°C. A declining slope indicates an ageing glass membrane. Logging both values at each calibration shows the sensor’s trajectory, so cleaning or replacement can be planned before the reading becomes unreliable, rather than discovered after a crop shows deficiency.
Choosing the right setup
For a hydroponic operation, the choice comes down to scale and maintenance philosophy. Smaller systems, or operators who prefer a swap-and-forget routine, are well served by the CPVC ST-710. Larger installations with multiple sensors, or sites where the lower recurring cost of replacing only the electrode head matters, are better suited to the ST-710SS with the EH-710 head.
Whichever is chosen, the principle is the same. In a nutrient solution, the reference junction is the part of the sensor most exposed to failure, and a larger salt bridge with more electrolyte behind it is what keeps the pH reading, and the dosing that depends on it, accurate for longer.
Autoflo Technology is the Malaysian distributor for Pyxis Lab sensors. For help specifying pH monitoring and dosing control for a hydroponic or fertigation system, contact us at info@autoflotechnology.com.