Library › Part F · Article 44 in the series · Level 1 · 11 min read

Meters and measurement: pH, EC, dissolved oxygen and temperature

Standfirst: After reading this article you will know what each of your meters actually senses, how to calibrate a pH meter and judge from its slope whether the electrode is worn out, how to correct an EC reading to 25 °C, and how to choose an oxygen sensor. You will also know how large a meter’s error is compared with your targets.

Before you read: pH, EC and concentration, Dissolved oxygen.

A meter measures something else and converts it

A pH meter measures a voltage, an EC meter measures how easily a current passes, and an oxygen meter measures a current or a light signal. Each turns that raw signal into a number through a calibration, and the calibration drifts as the sensor ages and gets dirty. This is why the reliability of a pH meter depends on how it is used and stored, and why it needs calibration with each use¹.

The pH meter: a voltmeter with a glass tip

A pH electrode ends in a thin glass bulb. When it is dipped in a solution, hydrogen ions exchange with sodium ions in a thin hydrated layer of the glass until the inside and outside are in balance, and this leaves a small voltage across the glass². The bulb is usually filled with a pH 7 buffer, so the electrode gives about 0 mV at pH 7, the isopotential point².

In theory the voltage changes by 59.16 mV for each pH unit at 25 °C (the Nernst slope), and by 55.18 mV at 5 °C and 64.12 mV at 50 °C². Measuring a voltage needs a second, steady electrode. In a combination electrode a silver wire coated with silver chloride sits in a salt filling solution, and a liquid junction, a small porous opening, lets that solution touch the sample and complete the circuit^2,3.

So the meter knows pH only through the conversion you set when you calibrate, and the circuit needs ions to carry current. In very dilute water, below about 100 µS/cm, the water does not conduct well enough to give a stable reading². If the pH reading of very pure raw water wanders, low conductivity is a likely cause (this article’s inference; see Water quality).

Calibrating a pH meter: offset and slope

Calibration fixes two numbers. The offset is how far the electrode’s voltage at pH 7 sits from 0 mV. The slope is how many millivolts the electrode actually gives per pH unit. You find both by measuring two or more standard buffers that bracket the pH of your samples². Buffers of pH 4.01, 7.00 and 10.01 are common². For hydroponic solutions, for which the suggested range is pH 5.8 to 6.5¹, the 4.01 and 7.00 pair brackets the range.

After calibration most meters show the slope as a percentage of the theoretical 59.16 mV². The US Geological Survey (USGS) treats 95% to 101% (56.2 to 59.8 mV per unit at 25 °C) as a properly working electrode, and below that it calls for cleaning, reconditioning or replacement². A laboratory-equipment maker accepts a wider 92% to 102% in a technical note⁴; the USGS range is the stricter of the two. The USGS also asks that the voltage in pH 7.00 buffer stay within 15 mV of the value the electrode gave when new².

The slope and the buffers’ pH both change with temperature, and meters with automatic temperature compensation correct for this during calibration². Calibrate with buffers within 10 °C of your sample, and after calibrating, read one buffer again: if it is more than 0.05 pH unit from its stated value, recalibrate².

Storing and distrusting a pH electrode

Keep the bulb wet in the storage solution the maker specifies, and do not store it in deionised water unless the maker says so²; one maker’s alternative is 100 mL of pH 7 buffer with 0.5 g of potassium chloride (KCl) added⁴, and a common grower’s pen ships with a KCl storage solution⁶. Rinse between buffers and samples, but do not rub or wipe the bulb, which the same maker warns can cause errors⁴.

Salt crystals in the reference junction are a common cause of failure². Reported working lives are normally 3 months, up to 1 year in special cases, in aquaculture use³, and about 12 to 18 months in environmental field use², depending on the water and on care. A slope that falls at each calibration shows the end is near (this article’s inference).

Paper strips and liquid colour kits use a dye that changes colour with pH; they are simple but less accurate³. A pH meter is usually accurate to at least 0.1 pH unit, while strips or colour standards may be enough for routine checks⁵.

The EC meter: two electrodes and a ruler

An EC probe holds two electrodes a fixed distance apart, and the meter measures conductance, the inverse of resistance, between them⁷. To get conductivity, which does not depend on probe size, it multiplies by the cell constant, the ratio of the distance between the plates to their area⁷. It uses alternating current, because a steady current would build a layer on the electrodes that changes the reading³; this is called polarisation.

Warm water conducts better because ions move faster (this article’s explanation). Conductivity rises by 0.5% to 3% per °C, depending on which ions are present⁷. Meters therefore correct readings to 25 °C, with either a linear factor α of 0.019 to 0.020 per °C or a nonlinear algorithm⁷. The linear form is:

EC₂₅ = EC_t ÷ (1 + α × (t − 25)), where EC_t is the reading at temperature t⁷.

Many meters compensate automatically⁸, and one tester lets you set α from 0.0% to 2.4% per °C⁹. The USGS states that compensation is typically accurate to within 5% between 5 and 35 °C for water of pH 4 to 11⁷. The meaning of EC itself is covered in EC and concentration.

Calibrating and caring for an EC probe

Most conductivity standards are KCl solutions, and one-point calibration commonly uses 500, 1,413 or 12,900 µS/cm; choose the one nearest your samples⁷. Behind them stand primary KCl standards: 0.01 mol/kg KCl conducts 1,409.3 µS/cm at 25 °C⁷. After calibrating, a check standard that reads within 3% of its stated value (above 100 µS/cm) needs no recalibration⁷. If the cell constant shifts by more than 2% between calibrations, suspect a contaminated standard or a dirty probe⁷. Fouling of the cell is a named source of error; clean it as the maker says, but never wipe or scour platinum-coated electrodes⁷.

A TDS or ppm reading is an EC reading multiplied by a conversion factor^1,10. One textbook uses 640¹, another describes 700 as usual¹⁰, one maker’s pen offers a 500 scale and a 700 scale¹¹, and another tester’s specification lists a factor of 0.45 to 1.00⁹. The same solution can therefore show different ppm on two meters. EC also cannot say which ions are present: an analysis of each element is more useful than EC for correcting a recirculating solution¹.

Dissolved-oxygen meters: membrane or light

Older oxygen probes are electrochemical¹², of the polarographic or galvanic type, as a sensor maker’s white paper groups them¹⁴. Oxygen diffuses through a membrane into an electrolyte around two metal electrodes, and a reaction similar to a battery’s makes the signal¹⁵. Polarographic probes also need the meter to power the sensor¹⁴. Because the reaction consumes oxygen, these probes need a constant flow of sample, as a sensor maker’s white paper explains¹⁴; without flow, the water at the membrane is depleted (this article’s inference). The USGS gives about 1 foot per second across the membrane (0.3 m/s, our conversion) and asks for hand stirring with hand-held probes¹³. The electrodes and electrolyte are slowly used up, and a new membrane takes about 24 hours to settle¹⁵.

Optical (luminescent) probes shine light on a dye whose glow oxygen suppresses; how long the glow lasts also depends on oxygen and gives the more reliable measure¹². They consume no oxygen and need no stirring¹². After calibration their stated accuracy is typically ±0.1 mg/L or 1%¹².

Makers recommend calibrating at 100% saturation¹², for example in water-saturated air: a few drops of water in the calibration cup, 5 to 10 minutes to settle, and the air pressure entered or read by the meter¹². A check should read within 0.2 mg/L or 2% of the computed saturation value¹². Temperature must be stable, because it controls how much oxygen water holds¹²; the solubility table is in Dissolved oxygen. Salinity lowers solubility a little: the USGS correction factor at 20 °C and 2,000 µS/cm is 0.9934¹², so for a nutrient solution of EC up to 2 mS/cm the salinity correction is under 1% (our calculation).

Temperature: three different numbers

The solution, the air and the leaf can all have different temperatures. Solution temperature controls how much oxygen the water holds¹² and enters the EC correction⁷, and it is often managed on its own: lettuce rafts are commonly chilled to 18 to 23 °C⁸. Air temperature at canopy height is the one used for humidity and vapour pressure deficit. A leaf can be warmer or cooler than the air: transpiration cools it, and in bright sun with partly closed stomata it can run 4 to 5 °C above the air¹⁶. An infrared thermometer reads leaf temperature, and a rise relative to the air is a sign of water stress¹⁷. The thermistors built into meters have their own error, ±0.5 °C on one tester⁹ and ±1 °C on one pen⁶. See Temperature, humidity and VPD.

Beyond meters: kits and laboratories

Colour kits test pH, ammonia, nitrite and nitrate with a few drops of reagent in a small sample, in five minutes or less; test strips are cheaper and only moderately accurate²⁰. In a Montana trial on soil extracts, nitrate strips tracked laboratory values (r² = 0.79), but their colour steps were coarse, and when five people repeated the test, the ratio of strip to laboratory values differed among them by a factor of up to 2.2¹⁹. The authors concluded such kits should not replace laboratory analysis¹⁹.

A laboratory analysis of a nutrient solution should include pH, nitrate, ammonium, phosphorus, potassium, calcium and magnesium¹. For micronutrients and for any diagnostic problem, a laboratory is recommended over kits¹. Ask the laboratory first how much to send and how; use a clean bottle, let the solution run for a few minutes, fill the bottle, empty it and fill it again¹. For slabs and bags, draw solution from the root zone with a syringe¹; in stone wool a simple suction device is standard¹⁸. For commercial hydroponic vegetables and cut flowers, weekly sampling with advice within two days is standard¹⁸. Leaf tissue sampling is covered in Nutrient disorders.

How accurate is accurate enough

A pH error of ±0.1 is small next to a 5.8 to 6.5 window^1,6. EC is tighter. A full-scale error of 2% on a 3,999 µS/cm range is about ±80 µS/cm (our calculation from⁹), which is ±6.7% of a 1,200 µS/cm reading. A lettuce raft operation that holds EC at 1.2 to 1.3 mS/cm⁸ works in a band 0.1 mS/cm wide, about the same size as the stated error of either EC meter in the table (±0.08 and ±0.1 mS/cm)^9,11. One reading cannot resolve that band; repeated readings on one regularly checked meter show the trend (this article’s inference).

Worked examples

(a) Correcting EC to 25 °C. Suppose a meter without compensation reads 1.30 mS/cm in a solution at 18 °C. With α = 0.019 per °C⁷: 1.30 ÷ (1 + 0.019 × (18 − 25)) = 1.30 ÷ 0.867 = 1.50 mS/cm (our calculation). With α = 0.020 the result is 1.51 mS/cm (our calculation). Resh’s table factor for 18 °C is 1.163⁸, giving 1.30 × 1.163 = 1.51 mS/cm (our calculation). The primary KCl standards give a check: 1,220.9 µS/cm at 18 °C and 1,409.3 µS/cm at 25 °C⁷, an implied factor of 0.0191 per °C (our calculation). The routes agree at 1.50 to 1.51 mS/cm, so the raw reading was about 13% low, and dosing until the raw number reads 1.5 would overfeed (our estimate).

(b) Judging a pH electrode. Suppose a meter last set to the ideal response (0 mV at pH 7, 59.16 mV per unit²) reads 4.10 in pH 4.01 buffer and 6.85 in pH 7.00 buffer, both at 25 °C. Converting the readings back to voltage (our calculation): (7 − 6.85) × 59.16 = 8.9 mV, and (7 − 4.10) × 59.16 = 171.6 mV. Using the USGS slope formula²: (171.6 − 8.9) ÷ (7.00 − 4.01) = 54.4 mV per pH unit, which is 54.4 ÷ 59.16 = 91.97%, about 92%, of the Nernst slope. Assuming the electrode gave 0 mV when new, the offset of 8.9 mV is within the 15 mV the USGS allows². The slope is below the USGS 95% limit² and just below the maker’s 92% limit, which the maker gives as 54.43 mV per unit⁴. Clean the junction, check the filling solution and recalibrate; if the slope stays below 95%, replace the electrode².

Keep a log

A record of results over time shows how each value changes and what corrections are needed¹. An FAO guide recommends a weekly logbook so trends can be followed²⁰, and the USGS logs every calibration’s slope and offset². A simple log has the date, pH, EC, solution temperature, water and nutrient added, and the last calibration slope; see Managing a recirculating solution.

Key numbers

Common mistakes

1. Storing a pH electrode dry or in deionised water. Electrodes are stored wet in a storage solution to keep their accuracy and response time, and the USGS advises against deionised water unless the maker allows it².
2. Calibrating once and trusting the meter for weeks. The slope and offset drift as the electrode ages, so calibration is advised with each use or daily^1,2.
3. Ignoring the slope figure after calibration. A slope below 95% points to a deteriorating electrode that needs cleaning, reconditioning or replacement².
4. Comparing ppm readings from two meters. Each maker uses its own factor, so the same solution gives different ppm^9,11.
5. Holding a membrane oxygen probe still in a tank. It consumes oxygen at the membrane, so its reading rises once the water is stirred^13,14.
6. Treating a test strip as a measurement. In one trial, strip results differed between readers by up to a factor of 2.2¹⁹.

Summary

Every meter turns a raw signal into a number through a calibration that drifts. A pH electrode gives about 59 mV per pH unit at 25 °C, and the slope reported after a two-buffer calibration shows whether it is still healthy². EC rises by about 2% per °C, so readings are corrected to 25 °C, and KCl standards set the cell constant⁷. Membrane oxygen probes need flowing water; optical probes do not. Solution, air and leaf temperatures differ. The stated error of an ordinary EC meter is about as wide as a typical EC target band, so a log of repeated readings on a checked meter is worth more than any single number.

Next in the series: Build: a single-container Kratky system.

References

1. Jones, J. B. Jr. (2005). Hydroponics: A Practical Guide for the Soilless Grower, 2nd ed. CRC Press, Boca Raton.
2. U.S. Geological Survey (2021). Measurement of pH. U.S. Geological Survey Techniques and Methods, book 9, chap. A6.4. https://doi.org/10.3133/tm9A6.4
3. Lekang, O.-I. (2013). Aquaculture Engineering, 2nd ed. Wiley-Blackwell, Chichester. Sections 24.3.3, Measuring pH, and 24.3.4, Measuring conductivity and salinity.
4. Thermo Fisher Scientific (2009). pH calibration procedure for optimal measurement precision. Orion technical note. https://assets.thermofisher.com/TFS-Assets/CMD/Product-Bulletins/TN-ph-calibration-procedure-for-optimal-measurement-precision-T-PHCAL-EN.pdf
5. Stickney, R. R. (ed.) (2000). Encyclopedia of Aquaculture. John Wiley & Sons, New York. Entry "pH".
6. Bluelab (n.d.). Bluelab pH Pen: technical specifications. Manufacturer product page. https://bluelab.com/products/bluelab-ph-pen
7. U.S. Geological Survey (2019). Specific conductance. U.S. Geological Survey Techniques and Methods, book 9, chap. A6.3. https://doi.org/10.3133/tm9A6.3
8. Resh, H. M. (2013). Hydroponic Food Production, 7th ed. CRC Press, Boca Raton. ISBN 978-1-4398-7869-9.
9. Hanna Instruments (n.d.). HI98129 low range pH/conductivity/TDS tester: specifications. Manufacturer product page. https://hannainst.com/hi98129-ph-ec-tds-tester.html
10. Mason, J. (n.d.). Commercial Hydroponics, 3rd ed. ISBN 978-0-9871022-2-5.
11. Bluelab (n.d.). Bluelab Conductivity Pen: technical specifications. Manufacturer product page. https://bluelab.com/products/bluelab-conductivity-pen
12. U.S. Geological Survey (2020). Dissolved oxygen. U.S. Geological Survey Techniques and Methods, book 9, chap. A6.2. https://doi.org/10.3133/tm9A6.2
13. Rounds, S. A., Wilde, F. D. and Ritz, G. F. (2013). Dissolved oxygen (ver. 3.0). U.S. Geological Survey Techniques of Water-Resources Investigations, book 9, chap. A6, sec. 6.2. https://pubs.usgs.gov/twri/twri9a6/twri9a62/twri9a6_6.2_ver3.pdf Superseded by reference 12; cited for the membrane-sensor procedure, which the 2020 edition no longer details.
14. Windsor, C., Steinbach, A., Lockwood, A. and Mooney, R. J. (2012). In-Situ optical RDO methods for dissolved oxygen measurements outperform traditional methods. In-Situ Inc. white paper. https://in-situ.com/pub/media/support/documents/Optical-Dissolved-Oxygen-Methods-Outperform-Traditional-Methods.pdf
15. Jenkins, T. E. (2014). Aeration Control System Design: A Practical Guide to Energy and Process Optimization. John Wiley & Sons, Hoboken.
16. Taiz, L. and Zeiger, E. (2002). Plant Physiology, 3rd ed. Sinauer Associates, Sunderland, MA.
17. Castilla, N. (2013). Greenhouse Technology and Management, 2nd ed. CABI, Wallingford.
18. Blok, C., de Kreij, C., Baas, R. and Wever, G. (2008). Analytical methods used in soilless cultivation. In: Raviv, M. and Lieth, J. H. (eds.), Soilless Culture: Theory and Practice. Elsevier, Amsterdam, pp. 245–289. https://doi.org/10.1016/B978-044452975-6.50009-5
19. Allison, E. and Jones, C. (2006). Accuracy of quick soil nitrate tests in Montana. Fertilizer Facts No. 39, Montana State University Agricultural Experiment Station. https://landresources.montana.edu/fertilizerfacts/documents/FF39QuickSoilNTest.pdf
20. Somerville, C., Cohen, M., Pantanella, E., Stankus, A. and Lovatelli, A. (2014). Small-scale aquaponic food production: integrated fish and plant farming. FAO Fisheries and Aquaculture Technical Paper No. 589. FAO, Rome. https://www.fao.org/3/i4021e/i4021e.pdf

This article was drafted with the help of AI tools and reviewed and edited by PonicLabs. Sources are listed above; tell us if you find an error.