ASA Lay Language Papers
161st Acoustical Society of America Meeting

An experience reducing toilet flushing noise reaching adjacent offices

By Noral D. Stewart, PhD, FASA, FASTM
Stewart Acoustical Consultants
7330 Chapel Hill Road
Raleigh, NC 27607

Presented Thursday afternoon, May 26, 2011
161st ASA Meeting, Seattle, Washington

Popular version of paper 4pAAb4

An office building was experiencing loud toilet flushing noise in a row of five offices adjacent to rest rooms on seven floors of the building.  In each row of five offices, the three in the middle were immediately adjacent to the rest rooms.  The middle office shared a wall with both the men’s and women’s rest rooms.  The toilets were a new design for very low water consumption, 1.2 gallon per flush.

An initial study by others documented maximum A-weighted slow sound levels in some of the offices greater than 60 dB with most greater than 50 dB.  “A-weighting” refers to an adjustment of the sound to account for the fact that the ear does not hear very low pitch and very high pitch sounds as well as sounds in the mid-range where most speech is.  “Slow” refers to a measurement method based on an average over a period of about a second.  The initial study recommended that the toilet sound be limited to 40 dB in the offices in accordance with guidance from the ASHRAE guide for mechanical system noise in buildings.  The background sound due to ventilation noise in offices is typically in the range of 35 to 45 dB.  Normal business conversation levels are typically around 60 dB.  A sound of 60 dB is typically judged to be 2 to3 times louder than a sound of 50 dB and 4 to 6 times louder than a sound of 40 dB.  However, physically, each 10 dB in sound level increase represents10 times as much sound energy present.  Thus, 60 dB is ten times the sound of 50 dB and 100 times the sound of 40 dB. 

The building owner made changes that were not fully documented and asked that the author repeat the measurements of the initial study.  On one floor a documented change was the addition of an extra set of studs and gypsum with a foam material in the cavity.  The repeated measurements showed that maximum levels had been reduced and were mostly in the range of 40 to 50 dB.  The levels in the two offices on the ends of each row of five, which were not actually immediately adjacent to the rest rooms, were below 40 dB.  It was interesting that on the floor with the added wall, the sound from the women’s rest room still exceeded 40 dB.   Maximum sound levels within the rest rooms were greater than 80 dB, and within the toilet stall of the loudest toilet a maximum level of 96 dB was measured.  This is about 20 dB louder than typical toilets.   Thus, if the toilets had been more typical, there probably would not have been a major problem.  There are no standards or requirements for manufacturers to report toilet noise levels, so it is hard for buyers to know when they are getting a noisy toilet.

The sound of the flushing had a strong high-frequency hissing sound.  This was in a frequency range where gypsum walls have a known weakness.  By comparing sound levels in the rest rooms and offices with the expected performance of the walls, it was determined that the sound in the loudest cases was coming through the walls.  One way to combat the transmission of the high-frequency sound is to avoid a rigid connection through the studs between the two sides of the wall and use a good sound absorbing material such as fiberglass in the wall cavity.  Where that is not done, new special gypsum panels are available with a damping agent in the middle of the panel that is especially good at combatting this weakness.  As a test, a layer of this damped gypsum was added to the walls in the three offices adjacent to the rest rooms on the floor where sound in the offices was loudest.  The maximum sound levels were reduced to below 40 dB in those offices.   

The remaining walls were then modified similarly.  The resulting average of the maximum level of 4 flushes was less than 40 dB in all offices, though some individual flushes were slightly greater than 40 dB.  One case where this occurred was adjacent to the women’s room where the wall had earlier been improved.  It was evident that structural flanking was now controlling the remaining sound reaching the offices in this and the other louder cases.  That is, the dominant sound in the offices is no longer the sound heard in the rest room coming through the wall.  Instead, the sound in the offices is due to direct contact of the toilets and their piping with the walls and floor.  This very efficiently introduces the sound into the structure.

The background sound in these offices from the ventilation system was extremely low, around 30 dB.  Thus, though the goal was achieved, the toilets could still be heard.  It would have been less noticeable with a more typical background sound.

It is important to recognize that some of the new toilets designed for low water consumption can be very noisy and that extra care is required in the building design when such toilets are used.  Consideration must be given to better walls, isolation of the toilets and piping from the structure with rubber or other soft material, and appropriate background sound from the ventilation system to mask or cover up remaining sound.