Church Acoustics
Church Acoustics
Wesely Logo

Wesley Music Centre

22 National Circuit, Forrest ACT 2603

Wesley Music Foundation

The Acoustics of Wesley Uniting Church

National Circuit, Forrest, ACT

 

This paper traces the changes effected in the acoustics of Wesley Uniting Church from 1983 to 2002.
It includes the results of acoustic tests from 1981 to 2003. While no tests provide a totally satisfactory assessment of the quality of an acoustic environment, they provide useful evidence along with what musicians and ordinary listeners can hear for themselves.

The authors hope that the lessons learned at Wesley will be of real value to those who are able and willing to take action to make any desirable changes in the acoustic environments of spaces of concern to them.

It should be no surprise that the improvements at Wesley have made a remarkable difference to the quality of sound for organ and choral music as well as congregational singing. The changes in the acoustics coupled with an organ rebuild and revoicing by George Stephens in 2000/2002 have delivered an organ sound today that bears little resemblance to that heard up to mid 1983.

Background

The Wesley Church was built in 1955. The architect and key church officials were concerned that the church be different from the nearby St. Andrew's Presbyterian, opened in 1934 (unfinished in relation to the original cathedral style), and with a well known resonance (often described in earlier days as “echo”) with poor speech characteristics. The St. Andrew's resonance was changed somewhat in the 1970s when a temporary wall was replaced by a largely glass extension. The Wesley architect followed trends in the 1950s which favoured non-resonant (lounge room) acoustics.

The ceiling in Wesley comprised sound absorbent “woodtex” tiles (straw impregnated with cement) with “perfotile” cladding across the main steel beams. These tiles have be used extensively in restaurants, offices and factories to deaden sound reverberations. The result was a dead and distorted acoustic with least reverberation in the lower and middle frequencies.

Fortunately the height of the ceiling (up to approx.10 metres in the nave) relative to the overall space provided potential for improvements.

Reports to Property Committee

When Garth became Director of Music in 1979 he raised his concern about the acoustics. He realised that the only way forward would be to research the issue, arrange acoustic tests and put forward a paper.

This was done in April 1981. The paper stated that

The Wesley acoustics are fundamentally deficient in relation to the length, amount and quality of resonance. This resonance in congregational singing with a full congregation is extremely low, to the extent that it would be almost non-existent to the average ear.
The key factor is the presence of sound absorbing material on the ceiling which is the principal means of transmission of sound energy in the church space.
The acoustic deficiencies are evidenced by

  • lack of warmth, sense of participation and vitality in congregational singing
  • inaudibility of unassisted speech
  • imbalance of tone, lack of transmission of tone, lack of tone quality for choral, organ and instrumental music
  • overall “telephone box” or “funeral parlour “ effect which is most marked with a sizeable congregation present

The Report drew on published articles by architects and sound engineers in the USA and UK, particularly in relation to the effect of sound absorbing material on church ceilings. Particularly useful were articles by B.G. Churcher in The Organ in October 1965 and January 1968, and also articles by David Klepper and Robert Newman in The Diapason in September 1970 and January 1971 respectively.

In October 1981 Wesley Property Committee arranged for Eric Taylor, an acoustic architect, to undertake tests using the standard equipment and procedure of the time. Reverberation times measured (albeit imprecisely by comparison with today's measurement standards) in terms of a decay of 60 decibels in the empty space were:

Reverberation times in seconds at various frequencies
Space 125Hz 250Hz 500Hz 1000Hz
Nave 1.0 1.5 1.8 2.3
Chapel 1.1 1.3 1.7 2.3

In his Report Eric stated:

The obvious initial priority, in order to increase the reverberation time of the interior spaces of the church, is to substitute an appropriate reflecting surface for the present “Woodtex” absorptive ceiling tiles and the perforated fibre tiles encasing the structural portal frames.

The ideal material for use in situations where organ and choral music is predominant, is polished timber in some form. Timber is noted for its warmth and resonance of tone and is highly recommended by authorities on church music. The three main points of consideration in the selection of the type of timber panelling are
(a) the effectiveness acoustically
(b) the manageability in terms of lightness and handling
(c) the comparative cost of construction

The recommended optimum material for the nave ceiling was veneered particle board having a thickness of 14.5mm including 1.5mm veneer on the exposed surface. It would be pre-cut and lacquered.
For the “perfotile” beams it was proposed that the perforations be filled using “Polyfilla” or patching plaster following by painting with two coats of coloured enamel or opaque polyurethane.

Eric also suggested that suspended and angled timber panels replace a concave ceiling above the choir area.

Property Committee Report and Decision by Congregation

At the end of 1982 the Chairman of the Property Committee put forward proposals to be discussed at a congregational meeting in February 1983.

The Chairman's paper canvassed differing views arising from a year's consultations on the acoustics and other matters.

The proposed work included

  • replacing the “woodtex” material with reflecting hardboard above the rafters plus insulating material under the metal roof (the roof to be removed thus eliminating the need for scaffolding in the church). The work was estimated to cost about $7,000.

  • as a lower priority, the rendering of the “perfotile”beams, reducing carpeting, re-angling of the ceiling over the choir area, adjusting the angle of the side walls of the choir, placing a cover over 60% of the false roof over the chapel and using portable, removable sound reflecting screens in the chancel area

Before the congregational meeting took place private funding was offered for the replacement of the “woodtex”ceiling.

At the congregational meeting in February 1983 overwhelming approval was given for the replacement of the “woodtex” ceiling in the nave. The only opposition came from those who did not want to see any change to the church or who thought any money offered should be spent on other things.

First stage

The removal of the “woodtex” ceiling in the nave and its replacement by reflective veneered particle board took place in mid 1983. It was done over two weeks. About half the ceiling was changed by the first weekend. The effect was dramatic. Those who opposed the change immediately acknowledged the effect on congregational singing and other music. The completed work was universally applauded.

Acoustic tests were conducted in July 1983. The results showed a major improvement in reverberation times at the lower and middle frequencies. By today's standards and methods of measurement the details are suspect and subject to wide interpretation. They are therefore not included here.

Second stage

In 1989 the “woodtex” tiles used in the ceiling at the front of the chapel were replaced by reflective veneered particle board. This made a significant improvement to sound reflection in the chapel area.

Third stage

In September 1991, the ceiling above the choir stalls and the choir and rear walls were changed to their present configuration. This resulted in a much improved sound egress from that area as well as an improved transmission of sound among singers within the choir stalls.

Fourth stage

Following a congregational meeting at the end of 1995, the following main works were undertaken in mid 1996 with about 80% of the funds being donated:

  • a complete rebuild of the floor in the chancel with replacement of a communion rail, some carpet and tongue and groove wood floor with a parquetry floor over sound proofing and the existing floor (the height was raised slightly) at a cost of approx. $ 33,500
  • a total covering of the “perfotile” cladding on the beams in the ceiling with plywood at a cost of approx. $ 17,000 (work on the cladding had become urgent as some had begun to be detached causing concern about potential falling tiles)
  • new church lighting

Work on the beams was undertaken using mobile scaffolding.

Fifth stage

After several months of experience, it appeared desirable to test the acoustics and explore the reasons for some distortions that were evident in recordings within the church especially with the organ.

The reverberation times were measured in November 1996 using a burst balloon method recorded by a RION type NA29E precision octave band sound analyser. The results were as follows:

  RT 60 in Secs at Octave Band Centre Frequencies in Hz
Configuration 125 250 500 1000 2000 4000 8000
Source @ rear choir 1.12 1.16 1.68 1.52 1.36 1.12 0.62
Receive @ organ console (towards  chapel)

The values of RT 60 in the direction between the choir and the chapel were deemed to be affected by early reflections of sound horizontally, which were received at approximately 80 ms after the balloon was burst in the choir stalls.

Although it was hard to find on the sound level meter, there must have been a secondary decay image in the RT 60 trace at a lower sound energy level, occurring at about 100ms after the balloon burst or 20ms later than the first primary image. These assumptions were based on the relative distances for the sound of the balloon burst to travel at 344m/s to the main reflecting surfaces in the chapel and to the rear of the nave respectively.

It appeared that the microphone located at the intersection of the nave and the transept between the choir and the chapel was picking up early and late reflections and therefore degrading the recording. A time difference of 30m/s or greater in the arrival of early and late reflections is usually perceived as an echo. Even though the time difference of 20m/s in the church was not quite distinguishable as an echo, it was long enough to cause a problem.

To remedy the situation, Eric Taylor proposed that the lowest frequencies be diffused by an array of pyramid panels in the chapel as follows:

Size of panel Frequency Equivalent Musical Note
1000mm square or rectangular 344 Hz Mid F (approx)
500 mm square or rectangular 688 Hz Treble F

These diffusing panels have been used many times. A notable example is in the Dallas Brooks Concert Hall in East Melbourne. At Attachment A are the designs used at Wesley in this fifth stage of acoustic improvements. The pyramid slopes were 5% and the MDF board thickness 20mm. The tender price was $1,950.

The panels were installed in February 1997. They made a further noticeable improvement in the acoustics, eliminating any detectable distortions due to the different echo times referred to above. They were useful in the diffusion of sound in the middle frequencies as evidenced in organ and choral music.

Sixth stage

Following another year's experience of the acoustics for various events, it was apparent to the trained ear and from recordings that increased diffusion in the higher frequencies would result in a better overall balance in resonance across the full range of Hz. Eric Taylor designed additional wall panels for the chapel beside the main chapel window and for the rear wall of the church. They were designed to provide diffusion of sound in the approximate range of frequencies from 500 Hz to 2000 Hz. The designs are at Attachment B. The 600 x 600mm panels were theoretically centred on 573 Hz, the 300 x 300 mm panels on 1147 Hz and the 150 x 150 mm panels on 2293 Hz. They are equivalent to D over three octaves. The pyramid slopes were 5%. The MDF boards were 10 mm thick for the smallest panels. The others were 20 mm. The tender price was $3,940.

The new panels were installed in June/July 1998. The results were very noticeable to musical ears with a marked further improvement in tonal balance and increased attenuation of sound over a wide range of frequencies. Many congregational and choir members as well as visitors commented on the overall quality of organ and choral sound reverberation with the church empty, partly full and full.

The following are the results of acoustic tests made on 11 August 2001 with an empty church:

Location of Measurement RT 60 at Octave Band Frequencies in Hz
  125 250 500 1k 2k 4k 8k
Source NE Chancel - Receive at Pulpit 0.96 1.70 2.04 2.02 1.70 1.50 1.16
Source in Choir - Rec. at Transept/Nave Inter. 1.28 1.96 2.04 1.99 1.60 1.39 0.76
Source Chancel steps - Receive rear of Nave 1.10 1.92 2.52 2.12 2.16 1.78 1.12
Source full organ C chord - Receive at Transept/Nave Intersection 1.36 2.08 2.52 2.24 2.28 1.82 1.20

Seventh Stage

In 2002, as part of a major property development on the site (including the construction of a Music Centre with a Music Room designed for practices, workshops, concerts and recordings seating 140 people), a gallery was built at the rear of the church. The front of the gallery is at about 2.5 metres above the floor of the church. It is 6.5 metres wide, conforming to the existing dimensions at the rear of the church. It consists of five levels about 900cm wide. At the lowest level it is 6.5 metres below the ceiling and at the highest level 5.5 metres below the ceiling. The floor is polished wood and it provides seating for 62 people on comfortable wooden chairs which are linked together. The chairs can be stacked or removed altogether for antiphonal choirs, orchestral instruments and/or electronic keyboard(s) for services and concerts.

Two sound diffusing 6 degree pyramids formerly on the rear wall of the church were placed on either side of the gallery between the gallery and the main church walls. Two larger 6 degree pyramids also from the previous rear wall were placed on either side of the large rear window with the bottom edges at average head height. The wooden front of the gallery consists of multiple pyramids of variable sizes. The ceiling under the gallery was designed to diffuse sound and comprised slopes of 8 degrees in the form of folded panels of plasterboard across the rear of the Nave.

The acoustic effects of the construction of the gallery with an empty church were measured on 28 April 2003. The results compared with those taken in 2001 before the gallery was built are as follows:

Location of Measurement RT60 at Octave Band Frequencies in Hz
  125 250 500 1k 2k 4k 8k
Source in Choir - Rec. at Transept/Nave Inter.2001 1.28 1.96 2.04 1.99 1.60 1.39 0.76
2003 1.28 1.80 2.08 2.08 1.84 1.68 1.24
Source Chancel steps - Receive rear of Nave 2001
1.10 1.92 2.52 2.12 2.16 1.78 1.12
Source Chancel steps - Rec. center of gallery 2003 1.32 1.84 2.24 2.20 1.92 1.80 1.36
Source full organ C chord - Receive at Transept/Nave Intersection 2001
1.36 2.08 2.52 2.24 2.28 1.82 1.20
Source full organ C chord -
Rec. center of gallery 2003
1.33 1.80 1.92 1.96 1.96 1.82 1.33

The results showed only fairly minor changes through the frequency range with a slight rise in reverberation time in the 4k and 8k octave bands with some variable changes up and down at other frequencies. To many who attend services and concerts who are conscious of sound quality, the consensus is that the overall acoustic has been enhanced through the greater diversity in sound diffusion with the gallery in place. The small difference in reverberation times over a wide frequency range contributes to overall sound quality and balance. The maintenance of good reverberation at the higher frequencies makes for clarity in sounds from upper voices, strings, organ and other instruments. It is interesting that the sound quality and balance between the different frequency bands is not noticeably different with a full church, notwithstanding a natural fall in reverberation times. There is a perceptible increase in reverberation at 125Hz and lower frequencies when the organ is played as compared with the pre-gallery situation. This is partly the result of increased wind pressures on the bottom octaves of several ranks (e.g. 16ft trombone and pedal diapason) in the final stage of the George Stephens rebuild in August 2002 after the gallery was installed.

Comparisons

While comparisons between results taken at different dates and with differing equipment are hard to assess the following summary of results may be of interest:

Nave RT 60 at Octave Band Frequencies (Hz)
  125 250 500 1k 2k 4k 8k
1981 prior to ceiling treatment
1.0 1.5 1.8 2.3 - - -
2001 after treatment and installation of diffusers
1.10 1.92 2.52 2.12 2.16 1.78 1.12
2003 after construction of gallery
1.32 1.84 2.24 2.2 1.92 1.80 1.36
Transept              
1996 before installation of diffusers in chapel/nave
1.1 1.2 1.7 1.5 1.4 1.1 0.6
2001 after installation of diffusers in chapel/nave
1.28 1.96 2.04 1.99 1.60 1.39 0.76
2003 after construction of gallery
1.28 1.80 2.08 2.08 1.84 1.68 1.24

In the case of the nave, the changes in reverberation times are not conclusive because of the limitations of the testing methods used in 1981, except in the octave bands of 500 Hz and below which show the effect of replacing the “woodtex” ceiling with timber panels.

However, in the case of the transept, increases in reverberation times between 1996 and 2001 clearly show the benefit of the diffusing panels on the walls of the chapel and the rear of the nave. These panels increase the length of travel of the sound waves before the sound energy decays, thereby effectively increasing the decay time. This diffusion has also enhanced the clarity of music and produced smoother decay curves, particularly at frequencies above 500 Hz.

Although not tested in the previous survey, the measurements in the nave with balloon burst and at the transept/nave intersection with full organ, showed an average reverberation time in mid frequencies in 2001 of approximately 2.5 seconds at mid frequencies. This figure was only slightly reduced with the construction of the gallery. The recommended reverberation time (RT 60) for the nave of a church of this volume (approximately 2,950 cubic metres) is within the range of 1.2 to 1.5 seconds when occupied by a 75% capacity congregation. Calculations indicate that the reverberation time with a 75% capacity congregation in Wesley Church is now of the order of 1.3 seconds at mid frequencies on the RT 60 basis and probably at least 1.5 for full organ. For each Hz band both the 2001 and 2003 results show remarkably smooth decay slopes compared to quite uneven decay in the results in the 1981 tests.

Finally, it should be emphasised that the standard definition for describing the reverberation time is the time taken in seconds for a sound abruptly stopped to decay by 60 decibels. However, let us consider the case of a full organ producing a sound level of say 100 dB (linear) in an empty church with a background sound level as low as 30 dB. It is possible therefore that someone with acute hearing could hear the sound decaying from 100 to 30, that is by an amount of 70 dB, which could add another 0.25 seconds to the perceived time of decay.

Lessons of experience

  1. If the acoustics of your church or favourite musical space leave much to be desired, research the problem and seek advice from an expert with a good track record of advising on new and old buildings.

  2. Take your time to explore optional solutions and be prepared for a long lead time to get support for appropriate changes.

  3. Be prepared to raise the necessary finance from private sources.

  4. Be prepared for opposition from those who can't detect a problem, who believe things must be left as they are or who say money shouldn't be spent on ' bricks and mortar'. Seek maximum support from all those involved in the music program or those who appreciate quality music. Be wary of acoustic “experts” whose agenda is to have an environment suited to microphones, loud speakers, electric guitars, drums and electronic keyboards, but not classical pipe organ, choirs and unassisted voice.

  5. Don't expect to get to the ultimate solution in one hit. It may take several or many stages, such as the seven stages at Wesley. At each stage, assess the results and plan the next stage as appropriate. In other words, “move as you prove” and practice patience. If you have success with the first stage, opposition to improvements tends to melt away. If it persists, you need to question in depth the objectives or motives of the opponents.

  6. At all stages, keep in mind the goal of the best possible acoustic environment for your situation. Length of reverberation time (especially for mid frequencies) is a key factor, but there is a range of factors which need to be taken into account in assessing the overall environment. These include the absence of any specific distortions, warmth, clarity, balance between frequencies, loudness, responsiveness, diffusion and the blend of sound. Don't be put off by uninformed comment, comment on one particular aspect or comment from particular individuals who may have a hearing problem or some axe to grind. Take opportunities to test the acoustics in different locations with a variety of instruments and voices and identify any enhancements that may be desirable. If church officials make changes that affect the acoustic environment (for example, they introduce new forms of seating or additional carpet, or add a gallery), be prepared for renewed action to retune the acoustics. If you hold to your long-term goal, the results can be most satisfying and even exciting.

Garth Mansfield OAM
Former Director of Music, Wesley Uniting Church
Forrest, ACT; Director of Wesley Music Centre, Forrest ACT,
Eric Taylor
Acoustical Consultant
Canberra, ACT

The above was published in OHTA News by the Organ Historical Trust of Australia in two articles - April 2002 and July 2004.

Garth Mansfield (1931- ) was Director of Music at Wesley Uniting Church, Forrest, ACT from 1979 to 2008. He has been Director of the Wesley Music Centre since 2002. He has been involved in various initiatives at Wesley Uniting Church including the commencement of the Wesley Scholarships Program in 1994 (70 organ, instrumental and voice scholars to 2009), the establishment of the Wesley Music Foundation in 1999, the comprehensive rebuilding of the organ by George Stephens in 2000/2002, the completion Wesley Music Centre contiguous to the Church in 2002 and the establishment of the ACT Organ School in 2003. Garth was awarded an OAM in 1994 for his services to church music.

Eric Taylor (1930-2009) was an acoustical architect, engineer and consultant from 1960. He provided advice on a wide range of projects including Wesley Uniting Church, a performing and recording space for the Wesley Music Centre and also music facilities at the RMC Duntroon and at various ACT colleges and schools. He provided successful acoustic solutions for auditoria (e.g. Dallas Brooks Hall in Melbourne), university halls, theatres, studios, churches and chapels.


---------------------------------------------------------

 

Enquiries and bookings: Garth Mansfield, OAM  Director of
Wesley Music Centre
Wesley Music Centre: Phone: 02 6232 7248
Fax: 02 6239 6836
Email

20 National Circuit, Forrest ACT 2603
(corner of National Circuit and Sydney Avenue)
PO Box 3217, Manuka ACT 2603