Almost every public description of a magnetic field reaches for tesla. The unit ampere per metre, symbol A/m, is the quieter partner in the same measurement chain, and it often shows up earlier than the tesla number that gets quoted. A Hall probe may read 0.5 mT inside a motor, but the engineer who designed the winding was working in amperes per metre long before the probe was brought anywhere near the air gap.
The distinction matters because A/m describes the magnetising field produced by current itself. Earth's field can be stated as roughly 40 A/m just as correctly as 50 microtesla at some locations; the two statements are linked by a constant. A/m also appears in coercivity specifications, core-loss data, and solenoid design equations written for engineers rather than for textbooks.
The ampere per metre is the SI unit of magnetic field strength, commonly called the H-field. It is a derived unit: the base unit ampere divided by the SI base unit metre. Its history begins with André-Marie Ampère and a tabletop experiment with two parallel wires in 1820, and its modern definition changed in 2019 when the ampere was redefined in terms of elementary charge. The practical effect of that change for most instruments was negligible. Its effect on calibration paperwork and on the value of the magnetic constant was not.
What A/m actually quantifies
H is a vector field that quantifies the magnetising influence created by free electric currents. In a long, tightly wound solenoid, the formula is short enough to write on a motor nameplate: if n turns are wound per metre and carry a current I, the H-field inside the coil is nI, expressed in amperes per metre. A coil with 100 turns per metre carrying 1 ampere produces 100 A/m. The formula does not require knowing what material is inside the coil.
That is Ampère's circuital law in its simplest form. The line integral of H around a closed path equals the free current enclosed. H is the field you get from the coil before the surrounding material responds.
Magnetic flux density B, measured in tesla, is the field that exerts force on moving charges and drives Hall sensors. In vacuum, B equals μ0H. Inside a magnetic material, B equals μ0(H + M), where M is magnetisation, itself measured in A/m. The shared unit is a source of confusion: H and M are both amperes per metre, but H is an external excitation field and M is a material response.
The practical distinction appears on a ferrite datasheet. One column lists H in A/m, the excitation. Another lists B in millitesla, the resulting flux density. A material that looks modest under one column can be extraordinary under the other, because a high permeability core turns a small H into a large B.
Two wires in 1820
André-Marie Ampère did not discover electromagnetism from a blank start. In July 1820, Hans Christian Ørsted announced that an electric current deflected a compass needle. Ampère moved quickly to reduce the phenomenon to a force law. He showed that two parallel conductors attract when their currents flow in the same direction and repel when the currents oppose. The force per unit length is proportional to the product of the currents and inversely proportional to the distance between the wires.
In 1827 Ampère published his Memoir on the Mathematical Theory of Electrodynamic Phenomena, Uniquely Deduced from Experience. The unit was named for him in 1881 at the International Electrical Congress, nearly half a century after his death. The ampere per metre we use today is a derived SI unit built from that name and an ordinary metre.
The ampere's definition changed in 2019 when the General Conference on Weights and Measures fixed the elementary charge e at 1.602 176 634 × 10-19 coulomb. A/m did not need a separate decision; it follows from the ampere and the metre. The NIST introduction to the ampere redefinition explains why the size of the ampere was preserved while its definition became quantum.
Photo by Logan Voss on Unsplash
Where the numbers sit in the real world
Earth's geomagnetic field at the surface ranges from about 25 μT to 65 μT in flux density. Dividing by μ0 gives an H-field of roughly 20 A/m to 52 A/m. A mid-latitude value around 40 A/m is common in order-of-magnitude discussions. Compass calibration and low-field magnetometer testing routinely work in both units.
The solenoid example returns with real numbers. A coil with 100 turns per metre carrying 1 A produces H = 100 A/m. In air, the corresponding B is μ0 × 100, which equals the old exactly-set value 4π × 10-7 × 100 = 1.2566 × 10-4 T, about 0.126 mT. Put the same coil around a core with relative permeability of 1,000 and the B field can approach 0.126 T, though saturation changes the real number. The coil did not change; the material did.
Magnet grades make the scale intuitive. Permanent magnet datasheets often list coercivity Hc in kA/m. Some commercial neodymium iron boron grades sit in the hundreds to thousands of kA/m. That number is the reverse field needed to bring magnetisation to zero, and it is a property of the material, stated in H units rather than B units by convention.
Conversion without losing the physics
The cgs unit oersted persists on older instruments and material sheets. One oersted equals 1000 divided by 4π amperes per metre, or approximately 79.577 A/m. In vacuum, 1 A/m produces about 1.2566 μT. A short conversion table is more useful than a paragraph of approximations.
| From | To | Multiply by |
|---|---|---|
| Oersted (Oe) | A/m | 79.577 |
| A/m in vacuum | Microtesla (μT) | 1.2566 |
| kA/m | A/m | 1,000 |
The vacuum conversion uses μ0. Before 2019, μ0 was exactly 4π × 10-7 henries per metre because the ampere was defined in terms of the force between wires. After the 2019 revision, μ0 is an experimentally determined constant. Its relative uncertainty is about 1.5 × 10-10, small enough to leave most engineering conversions untouched but large enough to matter in precision metrology.
Calibration: the gap between definition and reading
A definition is a promise. Calibration is the supply chain that keeps it. The SI Brochure can state the ampere exactly; it cannot tell a magnet producer whether a batch of Hall sensors drifts after a heat cycle. That is why reference magnets, field coils, and a chain of low-field calibrators exist in national metrology institutes and accredited laboratories.
When a laboratory calibrates a fluxgate or Hall probe, it generally compares the device against a known field source. Depending on the sensor, that source may be characterised as H in A/m or B in tesla. The current SI Brochure defines the units; the calibration procedure defines how faithfully a particular instrument realises them on a Tuesday afternoon.
Temperature, probe alignment, AC versus DC excitation, and even the screws holding a fixture can introduce errors far larger than the uncertainty in the SI definition. The gap between a standard in Paris and a reading in a motor test cell is not a scandal. It is the normal condition of measurement, and it is why engineering tolerances exist separately from metrological uncertainties.
Why the H-field survives in a tesla world
Every few years someone predicts that A/m will fade as devices read in tesla. It does not happen, because H connects directly to the ampere-turns in a coil. A motor designer computes magnetomotive force per metre before choosing a material. A transformer engineer uses H to read the B-H curve of a core, where the shape of saturation matters. A magnet manufacturer states coercivity in kA/m because the demagnetising field that kills a permanent magnet is best expressed in field-strength units.
Fluxgate and magnetoimpedance sensors, often used for low-field measurement, may report in nanotesla, but their calibration certificates frequently trace back to H-field standards. The dual vocabulary is not a failure of standardisation; it is a division of labour. B describes flux density and force. H describes current sources and magnetising demand. One cannot reduce the other, because the material in between changes the relationship.
The biography of André-Marie Ampère shows how quickly a laboratory observation became a defining quantity. The unit that carries his name is older than the telephone and the electric grid, yet it still does daily work in the oldest and newest electrical machines.
What changes next is not the definition
The next decade will not bring another redefinition of A/m. It will bring more places where the unit has to be trusted outside a metrology laboratory: electric vehicle traction motors, wireless charging pads, magnetic microsensors, and production-line quality checks. The hard question in year two after any new probe or standard is not whether the SI is correct. It is whether the reading on the factory screen matches the calibration certificate, and whether anyone tested the same sensor at 40 degrees.
The ampere per metre began as a force between two wires on a table. It became an equation, then a named unit, then an infrastructure of standard coils and calibration laboratories that most people never see. That is a longer and more useful history than the word tesla usually gets, even though tesla owns the headline.