Duct Silencer Insertion Loss and Pressure Drop
What a duct silencer costs in static pressure and what it actually buys in the room.
Example
You enter
- Airflow (cfm) 9000
- Silencer face width (in) 36
- Silencer face height (in) 24
- Catalogue pressure drop (in wc) 0.35
- at this face velocity (fpm) 1500
- Alternative face width (in, 0 to skip) 30
- Alternative face height (in) 20
- Target face velocity (fpm, 0 to skip) 1500
- Sound power upstream (dB, 0 to skip) 95
- Insertion loss (dB) 25
- Regenerated sound power (dB, 0 to skip) 72
- Fan available static (in wc, 0 to skip) 2.5
You get
- Face area (ft²) 6
- Face velocity (fpm) 1500 fpm across 6.00 sq ft
- Pressure drop 0.35 in wc
- Alt face velocity (fpm) 2160
- Squeezed into the smaller face 0.73 in wc at 2160 fpm (2.07x)
- Drop increase in wc 0.37576
- Attenuated lw db 70
- Downstream sound power 74.1 dB (attenuation alone would give 70)
- What another 10 dB of insertion loss buys 1.8587
Details, formula, and sources
Face velocity is the airflow through the silencer's own free area, and pressure drop rises with the SQUARE of it -- so a silencer squeezed into a duct that is already undersized costs far more than its catalogue figure suggests, and a 44 percent velocity increase roughly doubles the drop. INSERTION LOSS AND PRESSURE DROP ARE NOT INDEPENDENT: the geometry that absorbs sound, narrow passages and thick baffles, is the geometry that restricts flow. REGENERATED NOISE IS THE LIMIT PEOPLE DISCOVER LAST. At high face velocity the silencer generates turbulent noise of its own downstream of the baffles, and that noise is created AFTER the attenuation, so nothing removes it. Above some velocity, adding silencer length makes the system quieter on paper and no quieter in the room, because the regenerated noise has become the dominant source. Energy-summing the attenuated level and the regenerated level is what shows it, and what another ten decibels of insertion loss would actually buy against that floor is reported here, because that number is what settles the argument. Attenuation rises with LENGTH and so does pressure drop, but linearly; regenerated noise rises much faster with VELOCITY, and that asymmetry is the design rule. THE PRACTICAL SEQUENCE is to get the face velocity down by transitioning to a larger cross-section at the silencer, then choose the length for the attenuation needed, then check the drop against the fan's available static and the regenerated noise against the room criterion. A transition costs a fitting and buys back both the pressure drop and the noise floor. This scales an entered reference drop and energy-sums entered sound power figures. It does not read a manufacturer catalogue or predict insertion loss, pressure drop or regenerated noise from geometry, work in octave bands, distinguish forward from reverse flow rating (a silencer's rating differs with flow direction relative to the sound), model the duct system around it, or address breakout through the silencer's own casing. ASHRAE Applications, the silencer manufacturer's tested data, and the acoustical consultant govern.
face velocity = airflow / face area; pressure drop scales with the SQUARE of face velocity from an entered reference; the downstream sound power is the energy sum of (Lw - insertion loss) and the regenerated level.
Insertion loss, catalogue pressure drop and regenerated sound power are ENTERED from the manufacturer's tested data; a silencer's rating also differs with flow direction relative to the sound.
A square-law scaling and one energy sum.
Estimate. AHJ and licensed professional govern.
Field names used by the API: airflow_cfm, face_width_in, face_height_in, reference_drop_in_wc, reference_velocity_fpm, alt_face_width_in, alt_face_height_in, target_velocity_fpm, upstream_lw_db, insertion_loss_db, regenerated_lw_db, fan_available_static_in_wc, face_area_ft2, face_velocity_fpm, pressure_drop_in_wc, alt_face_velocity_fpm, alt_pressure_drop_in_wc, drop_increase_in_wc, attenuated_lw_db, downstream_lw_db, more_il_gain_db
- Catalogue data is entered insertion loss, drop and regenerated noise are tested valuesthe silencer manufacturer
- Square-law pressure drop scaled from one reference pointstandard duct fitting practice
- Direction not distinguished forward and reverse flow ratings differthe manufacturer's tested data