Home Nitrate Meters: What They Actually Measure

The device is the size of a phone. You push the probe into a tomato, wait a few seconds, and a number appears: 34 mg/kg. Green zone. The tomato is “clean.”

Here is the question almost nobody asks: where did that number come from? There are no reagents inside, no chromatograph, no nitrate electrode. There are two metal needles and a circuit board.

The answer exists, and it is not hidden. The manufacturers write down how the device works. That page simply goes unread, because nothing on it is exciting.

What the device actually measures

Here is the manufacturer’s own wording, from the Greentest site, one of the two most common brands:

The operating principle of the nitrate device GREENTEST is based on measuring the electrical conductivity of the fruit or vegetable.

Conductivity. Not nitrate content.

The needles send a weak current through the flesh and measure how easily it passes. The more dissolved salts in the sap, the higher the conductivity. An algorithm then converts conductivity into milligrams of nitrate per kilogram using a lookup table tied to whichever crop you picked from the menu.

So the device never sees nitrate. It sees the total quantity of ions and converts that total as though every ion were nitrate.

A claim you can check yourself

The manufacturer understands this and offers a justification. Verbatim again:

The analyses conducted in the certified laboratories showed that the component of nitrate ions typically account for over 98% of all conductive salts found in fruits and vegetables.

Everything rests on that sentence. If nitrate really supplies 98% of the conductivity, then measuring conductivity and calling it nitrate is legitimate. If it does not, the screen is showing something else under the label “nitrates.”

The convenient part is that the claim is testable at home. You need two numbers per vegetable: how much potassium it holds and how much nitrate. Both sit in open databases.

Potassium against nitrate

Potassium is the dominant cation in a plant cell. Those figures come from the USDA database. Nitrate comes from the 2008 EFSA opinion, which pooled 41,969 analytical results from twenty EU countries and Norway.

The comparison has to run on charge rather than milligrams: conductivity is carried by charged particles, so the count of charges matters, not the count of grams. Divide mass by molar mass and you get milliequivalents per kilogram.

The first row deserves a second look. In a median tomato, nitrate fails to reach even one percent of the charge carried by potassium alone. And potassium is not working alone in there: sodium, magnesium, calcium, phosphates, chlorides and organic acids all carry current too.

The claim was 98%. The result is 0.7% against a single item from that list.

Swap the median for the top 5% of samples and the highest figure in the table is beetroot at 67.7% of the potassium charge. Still nowhere near 98% of everything.

Why 98% is impossible in principle

Tables are optional here. Two pieces of high school chemistry are enough.

Charge balance. In any solution the sum of positive charges equals the sum of negative ones. Otherwise the solution would be electrically charged. Conductivity is carried by cations and anions together, both sides of the ledger. Nitrate is an anion, so it can claim half the conductivity at most. Even if every anion in a tomato were nitrate down to the last one, 98% would be out of reach.

Ion mobilities are similar. One could object that nitrate conducts far better than potassium and therefore “weighs” more in the reading. It does not. Limiting molar conductivity at 77 °F (25 °C): potassium 73.5, nitrate 71.4, chloride 76.3, sodium 50.1 S·cm²/mol. Potassium and nitrate are almost equal. Converting through charge is the correct move, and nothing is being smuggled in.

Which leaves one conclusion: the device measures a sum of salts and labels it nitrate. Inside a tomato, what it mostly measures is potassium.

We have covered this same principle before

Anyone who has run hydroponics knows all of this under a different name. There the instrument is called an EC meter, and we have a separate entry about it. The wording from that page:

EC measures the total concentration of dissolved salts in water.

No hydroponic grower ever says “my EC meter showed nitrogen.” They say “it showed 1.8 millisiemens,” and that means the sum of everything in the tank. How much nitrogen is actually in there gets calculated separately, from the recipe and the mass of each salt, and to confirm the fact the solution goes to a lab. That sum also includes whatever came out of the tap before the first gram of fertilizer, which is why hydroponics starts with testing the water itself, as the wiki explains in water analysis for growing.

A home nitrate meter is that same EC meter with needles instead of a beaker and a conversion table inside. The physics is identical. The only difference is that the screen prints the name of a specific compound.

The device asks which vegetable it is

Before each measurement the device asks you to pick a crop from a list. That is a required step of the method rather than an interface convenience: every crop has its own table converting conductivity into milligrams.

Two consequences follow. The obvious one: pick the wrong menu item and you get the wrong table and the wrong number. The second is more interesting. The device compares its figure against a limit and paints the screen green or red. In reality that limit is several numbers, not one.

Here is what EU maximum levels look like under Regulation 1881/2006, as reproduced in that same EFSA document.

For one and the same lettuce the limit swings between 2,500 and 4,500 mg/kg depending on the month of harvest and on whether it grew under cover or in the open. That is close to a twofold spread, and it was written in deliberately: less light in winter means the plant stores more nitrate, which is ordinary physiology rather than a violation.

The device in your hand knows neither the harvest month nor whether this head grew under plastic. It carries one limit per crop. Even with flawless nitrate measurement, the color of that screen would still be an approximation.

A cold tomato and a third of the reading

Measuring conductivity carries one more consequence, and the manufacturer warns about it directly:

Products should be measured at room temperature (at least 1 hour after refrigeration). Temperature deviations affect conductivity and nitrate measurement.

Let us put a number on “affect.” The conductivity of aqueous solutions climbs roughly 2% for every degree Celsius, about 1.1% per degree Fahrenheit, with the coefficient running 1.5% to 2.2% per °C for neutral salt solutions. That is exactly why laboratory EC meters compensate automatically back to 77 °F (25 °C).

A tomato out of the refrigerator sits near 41 °F (5 °C), a tomato off the counter near 72 °F (22 °C). The gap is 31 °F (17 °C). Multiply it out and the reading shifts by roughly a third. The nitrate inside that tomato did not move by a single milligram.

This is where those videos come from, the ones where somebody measures the same cucumber twice and gets two different numbers. Nothing is being faked. A thermometer is simply missing from the story.

How nitrate is measured for real

To have something to compare against, it helps to know which methods count as the real thing.

• Ion chromatography. The extract runs through a column where different anions travel at different speeds and leave it one at a time. Nitrate is measured once it has been separated from chloride, sulfate and everything else. This is the method used for official control.
• Colorimetry with cadmium reduction. Nitrate is chemically converted to nitrite, the nitrite forms a colored compound, and a photometer reads the intensity. The color is produced by a specific reaction of a specific substance.
• Ion-selective electrode. A membrane permeable mainly to the nitrate ion. Of the three, this is the only one that comes in pocket form.

All three share one step: nitrate is isolated from the other ions before anything is read. The home device skips that step entirely.

What a genuine nitrate electrode shows

Now for the interesting part. The ion-selective electrode is the correct method, and pocket instruments built on it do exist. How accurate are they?

A Spanish group from the University of Almeria tested exactly that across several seasons in Almeria greenhouses. A pocket LAQUAtwin against laboratory colorimetry, thousands of samples.

Look at the conditions. This is a research team recalibrating against standard solutions every two dozen readings, working with prepared samples, fully aware of what they are doing. Error of 14% in the best case, 22% in soil solution.

The manufacturer promised ±10%. Reality came out worse even in laboratory hands.

The sentence that lands on the home device

That same paper carries a conclusion worth reading slowly. The authors wrote that their system

cannot be recommended for the measurement of the [NO3-N] of undiluted petiole sap of vegetable crops.

Undiluted plant sap. Which is precisely the liquid a home device sticks its needles into.

The numbers behind that sentence: in undiluted sap the regression against the lab came out as y = 0.742x + 168, a systematic underestimate. Above 1,500 mg/L it collapses outright, with R² dropping to 0.389. The link between the reading and reality nearly vanishes exactly where the number is large, which is exactly where it matters most.

String the chain together. The correct method in expert hands is not recommended for undiluted sap. The home device works on undiluted sap, without calibration, without sample preparation, and without an ion-selective membrane.

The sensor wears out silently

One more part of the same study, and it is the most unpleasant one.

The researchers took 108 nutrient solution samples and read them twice with the same meter: first with a sensor worn down by ordinary use, then with a new one.

The worn sensor overstated by roughly 40%. The reading did not jump, did not rattle, did not throw an error. R² of 0.921 means the meter produced a smooth, tidy, stable line. Simply a displaced one.

That answers the question “what if mine is working fine.” There is no way for you to find out. A device with no reference to check itself against cannot report that it has drifted. It always shows a number, and the number always looks plausible. A hydroponic EC meter does get that check: it is tested regularly against a calibration standard of known conductivity, and the wiki explains how in its article on calibrating an EC meter.

What the manufacturers themselves say about accuracy

The two brands behave differently here, and the difference is telling.

Greentest publishes both the principle and the numbers. Stated range 0 to 9999 mg/kg, permissible error “less than 10%.” But the FAQ section carries a second formulation: accuracy “can reach up to 90%, typically around 77.5%.”

Sit with that. Accuracy of 77.5% is roughly ±22% error. By the manufacturer’s own account, the device performs worse than a professional ion-selective electrode in the hands of researchers. And that electrode, remember, is not recommended for undiluted sap.

The same pages list honestly what the device cannot do: it sees no pesticides, no insecticides and no heavy metals, “as these do not affect electrical conductivity.” The phrasing is wonderful, because it accidentally describes the whole design: the device sees whatever changes conductivity, and nothing beyond that.

SOEKS promises “the most accurate and sensitive probe for nitrates (nitrite) testing” on its store page, asks $199 for the EcoVisor F3, and gives no measurement principle, no error figure and no range. Same category of device, same promise, zero checkable numbers.

Why the reading still is not pure noise

Stopping at “the device measures the wrong thing” would be unfair. Some residual meaning is in there, and it deserves naming.

A heavily fertilized plant takes up more nitrate and more salts in general: nitrogen arrives as part of a nitrate salt, bringing potassium or calcium along with it. So a link between total conductivity and nitrate content does exist. It is indirect, and it is above zero.

That is why the device does not behave like a random number generator. Compare two heads of lettuce from the same shelf, at the same temperature, with the same crop selected in the menu, and the one reading higher does stand a fair chance of being “saltier” by total ions.

It breaks in three places at once. Across crops the potassium to nitrate ratio differs by an order of magnitude or more, so comparing a tomato with a lettuce is meaningless. Temperature shifts the reading by a third. And an indirect link never turns into milligrams per kilogram, which is precisely what the screen displays and precisely the figure that fails inspection.

Nitrite is a different substance

One of the manufacturers writes “nitrates (nitrite)” in a single breath, as if the two were interchangeable. They need separating.

Nitrate itself is relatively inert. The substance becomes a problem after reduction to nitrite, and nitrite is what stands behind the risks that got nitrate regulated in the first place. The quantities differ by orders of magnitude: where nitrate runs to thousands of milligrams per kilogram, nitrite runs to single digits.

Separating those two by conductivity is impossible in principle. Nitrite is also a singly charged anion of similar mobility, and against a background of potassium its contribution falls below what this kind of method can resolve at all. Laboratory protocols measure nitrate and nitrite through separate procedures for exactly that reason: they are two different jobs.

How much nitrate vegetables actually carry

If the device closes nothing, what remains is the boring route: knowing the numbers. Here are the medians from those same 41,969 EFSA analyses.

That table answers the question faster than any handheld can. The distance between a tomato and arugula is close to two hundredfold rather than a matter of percentages. No kitchen measurement bridges a gap that wide.

Where nitrate sits inside the plant, why the same arugula reads differently in different months, and what cooking does to it, are covered separately. What matters here is something else: most of the answer comes from the name of the vegetable, not from a reading.

How far the tails reach

A median is a midpoint, and it makes the spread within a single crop easy to forget.

The highest value in the entire EFSA database belongs to an oak leaf lettuce from Norway grown under cover: 19,925 mg/kg. That is four times the arugula median and more than four times the highest maximum level allowed for lettuce, 4,500 mg/kg.

For arugula itself, 56% of samples exceeded 4,500 mg/kg. For that crop, high nitrate is the ordinary condition of most samples rather than a rare anomaly.

Tails like those are what push people toward buying a device. The impulse makes sense: the numbers are real, the spread is wide, and none of it appears on the label. The trouble is how the error is distributed. A device measuring total salts does its best work on leafy greens, where nitrate makes up the largest share of conductivity. On a tomato, where nitrate is scarce, it is furthest from the truth. The check you need least is the one that works best.

Forty-seven grams of arugula

The acceptable daily intake for nitrate is 3.7 mg per kilogram of body weight, which works out to 222 mg per day for an adult weighing 60 kg (132 lb).

EFSA takes the arithmetic one step further, and the result is blunt: eating more than 47 g of arugula at the median concentration already crosses that line, before counting any other source of nitrate.

Forty-seven grams is a small handful, less than a standard clamshell off the shelf.

For context, the same document gives a baseline scenario: 400 g of mixed vegetables at median levels supply 157 mg of nitrate per day, comfortably inside the limit. So the deciding factor is which vegetable and how much of it, rather than vegetables in general. On that question, a device that confuses potassium with nitrate offers no help.

What to do instead

• Watch the crop, not the screen. Leafy greens and beetroot carry two orders of magnitude more nitrate than a tomato or peas. That is known in advance and needs no measurement.
• Watch the portion where concentration runs high. Arugula as a standalone dish on a large plate and arugula as a few leaves in a mix are two different things.
• If you need a real number, that means a laboratory. Ion chromatography or colorimetry on a specific sample. Slow and expensive, and it is a measurement of nitrate rather than a sum of salts.
• If the device is already bought, it has one honest use: comparing two samples of the same crop at the same temperature. Which of two heads of lettuce is “saltier” by total ions, it will show. Treating the absolute figure in milligrams of nitrate as a measurement is where it goes wrong.

What is missing here

I found no independent laboratory test of these particular consumer devices, from consumer organizations or in peer-reviewed journals. So this article contains no sentence of the form “model X is off by Y percent.”

What is established here needs no bench test of the device. The measurement principle is stated by the manufacturer. The 98% claim can be checked against two open databases, and it fails. Everything else follows from the physics of solutions.

The short version

A home nitrate meter measures the electrical conductivity of sap. The manufacturer says so. Calling that conductivity nitrate requires an assumption: that nitrate supplies over 98% of all dissolved salts. In a median tomato it supplies 0.7% of the charge carried by potassium alone.

The reading tracks the temperature of the product at roughly 2% per degree Celsius, so a tomato from the refrigerator and one from the counter differ by a third with the nitrate unchanged.

A genuine nitrate electrode, recalibrated every two dozen samples, carries a 14% to 22% error and is not recommended for undiluted plant sap. A worn sensor overstates by 40% without giving any sign.

The question the device was supposed to answer stays open. But most of the answer already sits in the name of the vegetable: from 1 mg/kg in peas to 4,800 in arugula.

A meter is not the only way to measure the wrong quantity. Folk honey tests do the same: water, napkin and match all read moisture. And the greenhouse versus field question is already settled in the limits: cucumber 300 against 150 mg/kg.

For heavy metals there is not even a flawed consumer device: lead is quantified by mass spectrometry or x-ray fluorescence. What follows for a kitchen is covered separately.