Key points
- Metal near a loop, mostly steel, weakens the field, and the high frequencies most.
- The loss depends on the metal, how much there is, how close the conductor runs to it and how wide each loop is.
- Find it on the site survey: building age and structure, drawings, and a look above the ceiling and below the floor.
- Allow for an overall drop (more current) and a treble drop (metal loss correction and voltage headroom).
- Layout choices can do more than a bigger driver.
- A design estimate is a plan; commissioning confirms the installed result.
What metal loss is
A hearing loop's current sets up an alternating magnetic field for the telecoils in hearing aids and cochlear implant processors. When that field passes through metal that conducts electricity, it induces circulating eddy currents in the metal. By Lenz's law, those currents set up a field that opposes the loop's own. Part of the loop's energy ends up as heat in the metal, and listeners get a weaker field. Steel is magnetic too, so it also pulls the field out of shape.
On site, metal loss shows up three ways:
- A weaker field overall, so the loop needs more current to reach the target level at listening height.
- Duller sound, because the high frequencies drop further than the midrange.
- Local dips over concentrated steel, such as a beam line or a steel stage frame.
Why the high frequencies suffer most
The voltage induced in metal depends on how quickly the field changes, and a 5 kHz signal reverses direction five times as often as a 1 kHz signal. Higher frequencies drive stronger eddy currents, so the loss grows with frequency. Set the level at 1 kHz, and a loop over a steel deck can read correctly there yet still be several decibels short at 5 kHz, while at 100 Hz the loss is usually small.
Consonants such as s, t, f and sh carry much of their energy in the upper frequencies, so listeners often describe uncorrected metal loss as loud enough but muffled.
The frequency-response target in IEC 60118-4 asks that the field from 100 Hz to 5 kHz stay within 3 dB either side of its 1 kHz level (the targets explained). Metal loss is a common reason a loop misses it. A driver that runs short of voltage as the loop's impedance climbs toward 5 kHz makes it worse (sizing a hearing loop driver).
Where the metal hides
| Construction | Where the metal is | Typical effect |
|---|---|---|
| Timber frame, masonry, unreinforced concrete | Little or none | Low |
| Water underfloor heating in plastic pipe | Mesh in the screed or a metal manifold, not the plastic pipe | Low, unless reinforced |
| Suspended ceiling grid, metal tiles, ducts | Near a ceiling loop | Low to moderate; more when the loop lies right on the grid or tiles |
| Reinforced concrete slab (rebar or mesh) | Under a floor loop | Moderate to high; worse with dense or double mesh |
| Steel frame (beams, columns, joists) | Above or below the loop | Local dips along beams; depends on distance and direction |
| Heavily reinforced or post-tensioned slab | Dense rebar, steel tendons | High |
| Steel raised access floor | Steel panels right under the loop | High to severe |
| Steel composite deck (corrugated steel under concrete) | A continuous steel sheet under the floor | High to severe |
These ratings are typical. The real loss depends on the steel in the building, so measure on site when it matters.
Distance and loop size can matter as much as the material: a conductor in or on a reinforced slab loses more than one raised away from it. Wide loops generally lose more than narrow ones over the same steel, and a conductor running alongside a beam loses more than one crossing it at right angles, because a parallel run couples into the steel along its whole length. Metal that forms a closed ring, such as a ceiling grid with electrically connected joints, carries eddy currents more easily than metal with breaks in it.
Spotting it on the site survey
You can't see rebar through a finished floor, so collect clues. The hearing loop site survey checklist has the full list; for metal, focus on these:
- Ask about the structure. When was it built, is it steel-framed or concrete, and what is under this floor? Ask for structural or as-built drawings. If the slab may be post-tensioned, find out before anyone cuts or drills it: a damaged tendon is dangerous.
- Look from below. From the floor beneath, a corrugated steel deck is easy to recognize. Where a floor has been cut for a floor box or trench, look at the edge for bar or mesh.
- Look above and lift a panel. Under a ceiling tile, note the grid, ducts and any steel deck above; an access floor panel shows whether the floor is steel.
- Use simple tools. A magnet sticks to steel studs, deck and grid, though not to aluminum. A rebar scanner shows whether a slab holds bars, roughly how far apart they are and how deep.
- Mark local steel on your sketch: beam lines, steel stages and risers, metal seating frames, steel stairs.
- Record the conductor height: in the slab, on it under carpet, or in the ceiling void, and how far that is from the nearest metal.
Where the construction is unknown or clearly heavy, many installers measure the loss with a temporary test loop before committing to a design. That procedure is outside the scope of this article.
Allowing for it in the design
The overall drop: more current
Field strength rises in step with current, so a loss in decibels converts straight into a current multiplier:
| Loss at 1 kHz | Current needed, compared with no metal |
|---|---|
| 2 dB | × 1.26 |
| 4 dB | × 1.58 |
| 6 dB | × 2.0 |
| 10 dB | × 3.16 |
Doubling the current takes twice the voltage and four times the power into the same loop, so heavy loss soon decides the size of the driver.
The treble drop: correction and headroom
Many loop drivers have a metal loss correction (MLC) control that lifts the high frequencies to restore the response. It asks for more current at the top of the band, where the loop's impedance is highest, so the driver needs voltage in reserve. Without headroom at 5 kHz, the driver can clip on sibilants once the correction is set.
Layout and conductor
Before reaching for a bigger driver, look at the layout:
- Add distance. A loop in the ceiling void over a steel floor may do better than one on the floor, if the ceiling holds less metal and its height suits the listeners. See listening height.
- Use narrower loops. A phased array of narrow elements usually holds up better over steel than one wide perimeter loop, and keeps the field more even.
- Cross the steel. Run conductors across beam lines rather than along them where the room allows.
- Watch the turns. More turns lower the current but raise the inductance, so the driver needs more voltage at 5 kHz. Choosing copper tape over round wire changes the voltage far less; see loop wire vs copper tape.
IEC TR 63079, the IEC's code of practice for hearing loop systems, includes a preferred layout for steel-framed buildings and ways to break eddy-current paths in metal structures.
In Hearing Loop Designer, you pick the construction type and a low, typical or high estimate, and the modeled field strength, frequency response and driver current and voltage include it, so you can compare layouts before anything is ordered (metal-loss planning).
A worked example
Community Hall is 18 m × 12 m, with seated listeners at 1.2 m and a perimeter loop under carpet on a reinforced concrete slab.
- Modeled as a single turn with no metal, the loop needs about 6.5 A at 1 kHz to reach 400 mA/m at listening height in the middle of the room.
- The survey points to a standard reinforced slab. Take an estimate of 4 dB loss at 1 kHz and 11 dB at 5 kHz.
- The overall drop: 6.5 A × 1.58 is about 10.3 A at 1 kHz.
- The treble drop: 5 kHz sits 11 − 4 = 7 dB below the 1 kHz level, well outside the ±3 dB window. Bringing it level with 1 kHz takes about 7 dB of lift: 2.24 times as much current at 5 kHz as at 1 kHz, or about 23 A for a test tone at full level, at the frequency where the loop's impedance is highest. Speech carries much less energy at 5 kHz than at 1 kHz, so the driver doesn't need to hold 23 A continuously, but the correction does use up voltage where the loop needs the most. Check it as described in sizing a hearing loop driver.
- If the planned driver can't deliver that, change the layout first and model it again.
Confirm it at commissioning
A metal loss figure in a design is an estimate: construction varies and drawings are not always as built. Commissioning measures the installed loop across the listening area at listening height. When the level is checked with a wideband test signal, IEC 60118-4's measuring method sets the order: background noise first, then the driver adjusted for as flat a frequency response as you can get, and only then the reference level. It warns that skipping that step can give large errors, particularly in rooms with metal reinforcement. See commissioning a hearing loop.
Common questions
Does aluminum cause metal loss too?
Yes. Aluminum is not magnetic, but it conducts electricity very well, so eddy currents still form in it. A magnet won't find an aluminum ceiling grid, cladding or floor panel, so look and ask as well.
Can I fix metal loss by turning the driver up?
Partly. More current makes up the overall drop, but the treble stays low compared with the midrange, so speech still sounds dull. Restoring it takes metal loss correction and driver headroom. Turning up also can't fill a local dip over a steel beam, and a stronger field carries further beyond the room.
Will the estimate in the design match the finished room?
Treat it as a planning figure, not a promise. Commissioning measures what the installed loop delivers, and the driver's correction is set from those measurements. If the survey left real doubt about the construction, note it in the design and allow extra driver headroom.
Sources
- IEC 60118-4:2014+AMD1:2017 CSV, Electroacoustics - Hearing aids - Part 4: Induction-loop systems for hearing aid purposes - System performance requirements. International Electrotechnical Commission (IEC). Read October 4, 2026.
- IEC TR 63079:2017+AMD1:2018+AMD2:2020 CSV, Code of practice for hearing-loop systems (HLS). International Electrotechnical Commission (IEC). Read October 4, 2026.
- Best Practices for Hearing Loop Installation. Hearing Loss Association of America (HLAA). Read October 4, 2026.
- Faraday's Law (HyperPhysics). Georgia State University, Department of Physics and Astronomy. Read October 4, 2026.
We describe IEC 60118-4 in our own words and cite the source for every fact. This is general information for installers, not advice for a particular building. Spot something out of date? Write to dave@equalaccessaudio.com.