Commissioning a hearing loop: what happens after the design

A design tells you what a loop should do. Commissioning finds out what it does: you measure the installed system, adjust the driver and write the results down for the venue. Here are the steps in the usual order, with the IEC 60118-4 targets in our own words. For the numbers behind them, see IEC 60118-4 field strength: 400 mA/m and ±3 dB explained.

Key points

• Noise first. Background magnetic noise is measured with the loop off, before you adjust anything.
• Set the driver by its maker's procedure: level, then treble correction, then the level again.
• Measure where people listen: across the agreed listening area at head height, usually 1.2 m seated and 1.7 m standing.
• Judge treble against 1 kHz. From 100 Hz to 5 kHz the field should stay within ±3 dB of the 1 kHz reading at the same spot.
• Finish with people, then write it down. A hearing aid user tries the loop, and the record goes to the venue.

What to bring

• A field-strength meter for loop work (A-weighted for noise, flat for signal), with its calibration date.
• Test tones at 100 Hz, 1 kHz and 5 kHz, and a loop listening receiver.
• A stand to hold the meter at listening height, the design, and the record sheet.

Before you power up: check what was built

With the driver disconnected, measure each loop's resistance and insulation to earth, and compare them with the design and your readings from before the floor was covered. Check the routes against the drawings, using photos taken before the floor was covered: an array spacing that changed on site shows up later as a weak stripe. Mark changes on the plan so the record shows the loop as built.

Background magnetic noise

Turn the driver off and measure the A-weighted noise across the listening area at listening height, with the building running normally: lights at several dimmer settings, heating and ventilation, lifts and stage equipment. Listen with the receiver too; a dimmer buzz can bother listeners more than a steady hum at the same reading.

Readings are in dB(A) re 400 mA/m, so they sit well below zero. IEC 60118-4 treats −47 dB(A) as the ideal: noise that low should go unnoticed. It accepts a higher level as the normal limit, and a higher one again only for short periods of use.

Ideally noise was first measured on the site survey; commissioning confirms nothing has changed. If it is high, find the source, because listeners will hear it as hum in the loop. Wiring faults can create a strong field, and fixing them is a job for a licensed electrician.

Worked example: noise in Community Hall

Illustrative numbers, synthetic room. With the loop off, most of the hall reads about −50 dB(A). Beside the dimmer rack it reads −41 dB(A), 6 dB above the −47 dB(A) ideal though inside the higher limit, and the receiver picks up a buzz that changes as the lights dim. Record both readings, mark the corner on the plan, tell the client, and ask the venue's electrician to check the dimmer wiring.

Setting the driver

Follow the driver maker's procedure. Drivers differ in how input gain, automatic gain control (AGC) and metal-loss correction interact, and a control marked "0 dB" does not mean the field reads 0 dB. A common sequence:

1. Input. Feed the normal program source, or the maker's test level, and set the input gain so the AGC works in its normal range.
2. Level. With a 1 kHz tone and the meter at a reference position at listening height, bring the output close to 0 dB re 400 mA/m. The steady tone stands in for the loud moments of speech.
3. Treble. Play 5 kHz at the same input level and use the metal-loss correction to bring it within ±3 dB of the 1 kHz reading. Check 100 Hz the same way.
4. Set 1 kHz to 0 dB, because treble correction can move it.

Watch the clip or limit indicators. A driver that reaches 0 dB at 1 kHz but runs out of voltage at 5 kHz can sound dull or distorted on loud speech; see Sizing a hearing loop driver and Metal loss in hearing loops. For a phased array, check each channel on its own for faults, then set the level with both running.

With the driver set, stop the program, switch the microphones off and measure the noise again at a few points. The reading should barely move. A clear rise points to hum or hiss from the driver or the audio feed, such as a ground loop in the sound system. As a rule of thumb, a rise of more than about 3 dB is worth tracking down.

Field strength across the listening area

Agree the listening area with the client and write it down: seats, wheelchair spaces, standing areas, lecterns or counters. The strip right next to the wire usually sits outside it. Then measure:

• At listening height. Listening height for seated and standing listeners covers tiers, stages and balconies.
• With the meter upright, reading the vertical part of the field that a telecoil picks up.
• On a grid close enough to catch a dip, plus the hard spots: the middle of wide rooms, corners, steel columns, the front row and the edges. In a phased array, run lines of points across the loop segments, not only along them, because a weak stripe runs the same way as the wires.
• With the same test signal at every point, writing each reading on the plan.

Every position should read within ±3 dB of 0 dB. Check the spread first (highest minus lowest, 6 dB at most), then the level.

Worked example: a 300-seat sanctuary

Illustrative numbers, synthetic room. A two-channel phased array under the pews is measured at 1.2 m at 30 points. Twenty-eight read between −1.8 dB and +1.9 dB; two beside a steel column read −3.9 dB and −4.2 dB.

The spread is 6.1 dB, so the level controls cannot fix it: raising the output 1.2 dB lifts the low seat to −3.0 dB but pushes the high one to +3.1 dB. Check whether those seats are in the agreed area and look for a route change near the column. Otherwise record them as exceptions agreed with the client, rather than turning the whole loop up.

Frequency response

At a few positions (the reference point, over the most steel, and the far end) measure 100 Hz, 1 kHz and 5 kHz at the same input level, or use pink noise if that is the maker's method. Judge 100 Hz and 5 kHz against the 1 kHz reading at the same spot, not against 400 mA/m.

Illustrative numbers, same sanctuary. Over the steel deck: −0.6 dB at 100 Hz, −0.2 dB at 1 kHz, −2.9 dB at 5 kHz. Against 1 kHz, that is −0.4 dB and −2.7 dB: inside ±3 dB, but close at 5 kHz, so note it and check the driver on loud program.

Neighboring rooms

If the room next door has its own loop or needs privacy, measure there too: by the shared wall, above and below, with your loop running and theirs off. Record where you measured and what you heard. See Hearing loop overspill for layout options.

Listening checks

A meter does not hear distortion or a badly mixed feed. Walk the room with a receiver while each microphone and playback source plays. Listen for steady level, clean speech on loud passages, hum, dropouts, and the right mix (speech microphones, not room ambience).

Then ask a hearing aid or cochlear implant user who knows loops to try it on their telecoil program while you are still on site, as the Hearing Loss Association of America (HLAA) recommends. Check the signs are up and staff know how to switch the system on.

The commissioning record for the venue

The record is what the venue keeps and what anyone servicing the loop later compares against. It records what was measured, not a promise about every hearing aid. IEC TR 63079, the IEC code of practice for hearing-loop systems, gives fuller guidance on testing, operation and maintenance.

Add when the loop should next be checked, and note that new flooring or ceiling work can change the field. Sign and date it.

Where Hearing Loop Designer fits

Hearing Loop Designer prints a commissioning record from the design. The project details, driver and listening height are filled in and the planned loop is drawn on the sketch, with room to write in the measurement points, background noise, field strength and frequency response before and after adjustment, and the listening check. Your site readings then sit on the same sheet as the plan they test. The software checks the design; installed performance is confirmed by commissioning.

Common questions

Can I commission from the design report?

No. The design predicts the field, but it cannot see hidden steel, background noise, the as-built route or the final driver settings. Those show up on site.

Do I need to measure every seat?

No. Measure a grid close enough to catch a dip, plus every awkward spot: corners, columns, the middle of wide rooms and the edges of the listening area.

Why use a hearing aid user as well as a meter?

A meter shows level and frequency response. A listener hears distortion, noise and how natural speech sounds through a telecoil.

How often should a loop be rechecked?

Agree an interval with the venue and write it in the record; HLAA lists periodic maintenance among its best practices. Recheck after building work near the wire too. A quick listen with a receiver before each event catches most faults early.