Notification Sound Level Fails the Acceptance Check
A machine workshop, 30 m × 18 m with a 6 m ceiling, has been fitted with eight horn/strobes on a 70 V line. During acceptance testing you measure 89 dBA in the far corner while the machines are running; a background reading of 62 dBA was used at design stage. The installer says the design target was 62 + 15 = 77 dBA, so 89 dBA is a comfortable pass. The horns are mounted at 4.5 m, tapped at the lowest wattage setting, and two of the eight are aimed into a mezzanine rather than across the floor. A separate reading 1 m from one horn reads 106 dBA. The workshop operates three shifts and the owner wants the system signed off that afternoon.
1. Determine the required sound level and judge whether the installation passes.
Read the worked answer
- Required level is based on the normal ambient with equipment running, not a quiet-condition reading: 88 dBA ambient measured during production.
- Required: 88 + 15 = 103 dBA, comfortably below the 105 dBA upper limit for occupied areas.
- Measured worst-case: 89 dBA - 14 dB short.
So the installation fails, and the installer's calculation fails for a more basic reason than arithmetic: it used a 62 dBA ambient that does not represent the space in use. The 106 dBA reading near a horn also matters - the fix must raise the floor level without pushing any occupied position over the upper limit.
Note that a 14 dB shortfall is not a small trim. With roughly 6 dB per doubling of distance or per doubling of appliances, closing 14 dB needs substantial redesign, not a tap adjustment alone.
2. List the likely causes and rank them by how much they can realistically explain.
Read the worked answer
- Design basis error (largest): ambient measured with machines off. Every downstream number inherits the error - 26 dB of ambient understatement.
- Tap setting: all horns at minimum wattage throws away output the amplifier can already deliver. Cheap to fix, and it is the first thing to try, but it will not close 14 dB by itself.
- Aiming and height: horns at 4.5 m directed at the mezzanine leave floor zones with no direct path; sound is being placed where people are not.
- Coverage gaps: two of eight appliances effectively serve the mezzanine, so the floor has six.
- Absorption/obstruction: machinery and stock absorb and shadow sound; the far corner is both furthest and most obstructed.
Ranking matters: adjusting taps on a layout that is aimed at the wrong place produces a passed meter reading with poor coverage in the occupied zone.
3. Describe the re-test protocol you would insist on before sign-off, including what you record.
Read the worked answer
- Test with the building in its normal operating condition - machines running, doors in normal position.
- Use an A-weighted, slow-response meter, held at roughly 1.5 m above the floor at the listener's ear position, at the worst-case points, not next to an appliance.
- Record both the average ambient and any maximum ambient lasting more than about a minute, since requirements are stated against both.
- Record the measured level at each test point, the required level, and the margin.
- Check that no occupied position exceeds the upper limit - particularly near horns after any tap increase - and confirm audibility against expected noise peaks, not just steady machinery.
- Repeat after any change to taps or placement; a failed reading means redesign, so the record must show the final configuration, not the first attempt.
Reasoning: the number is only meaningful alongside the test conditions. A reading taken in the wrong condition is how this job went wrong the first time.