ACOUSTICAL DESIGN PRACTICUM

Peabody East Hall Sound System Analysis and Design

 

 

 

 

 

HADI SUMORO

 

CONTENTS

I.                   Abstract………………………………………………………………………… 2

II.                Introduction……………………………………………………………..………  3

·         Peabody East Hall Acoustical Data………………………………...………   4

·         Preparation and Equipments………………………………………..………  6

III.             Stage Analysis………………………………………………………….………   7

·         Auralization………………………………………………………………… 8

·         Analysis…………………………………………………………….……… 9

IV.             Installed Sound System Analysis………………………………………………     10

·         EASE Area Mapping……………………………………………….……… 10

·         PAG-NAG Calculation……………………………………………..……… 13

·         Analysis…………………………………………………………….……… 14

V.                New Sound System Design…………………………………………….………   16

·         EASE Area Mapping……………………………………………….……… 16

·         PAG-NAG Calculation……………………………………………..………  19

·         Approaches, Consideration and Analysis…………………………..………   20

VI.             Additional Acoustical Recommendations………………………………………     22

VII.          Conclusion……………………………………………………………...………  24

·         EASE Area Mapping……………………………………………….……… 24

·         Comparison………………………………………………………………… 29

VIII.       References………………………………………………………………………  30

IX.             Documentation………………………………………………………….………   31

I.                   Abstract

 

Peabody East Hall is often used to accommodate jazz performances like in nightclubs and used as rehearsal hall for classical performances. The need of a sound system which can recreate a real life club situation is the focus of my study. Several approaches have been taken such as surveys, acoustical testing and computer modeling to design an appropriate sound system and to determine a good location for the stage.

 

 

II.                Introduction

 

Peabody East Hall is a hall that holds rehearsals and/or performances that range from small ensemble, jazz big band or jazz/classical orchestra. There is no fixed stage and audience area. However, a sound system is installed and this forces a stage area to be located in a certain fixed place when the installed sound system is used. Surveys (interview with students and staff), analysis and a site survey were done to analyze the performance of the sound system in the Peabody East Hall. Several considerations for this hall are:

·         The need to create a nightclub/jazz club atmosphere especially for jazz students.

·         A low-mid frequency resonance is detected on the stage position under the installed sound system.

·         The studio recording contains noticeable sound leakage from the loudspeakers and degrades the recording quality.

·         Maximizing the effective audience area (increasing the uniformity of the loudspeaker coverage area).

A stage and installed sound system analysis were performed and completed. A new sound system design and relocation of the stage are also considered in this report. Acoustical analysis is not focused but some recommendations are made. Several approaches taken to support the analysis and design are:

·         Computer modeling.

·         Acoustical testing and data analysis.

·         Room interpretation/evaluation using convolution method.

Peabody East Hall Acoustical Data

The approach of the analysis and design is focused on how East Hall is usually used. The acoustical draperies are normally extended to limit reverberant sound field built up and this report is based on this condition. The reverberation data of the Peabody East Hall with the draperies extended are shown below:

The reverberation measurement was done in early spring 2007 semester. The data were taken using EASERA, Earthworks M30BX, M-audio Transit USB soundcard and 15” balloon as the impulse source. The data shown above has been averaged from several T20 and T30 measurements. The background noise level was measured using Norsonic 118 SPL meter and the data is shown on the following page.

Peabody East Hall has a rated NC value of 19.

 

Preparation and Equipments

·         Computer Model

A wire-frame computer model of the Peabody East Hall has been developed using EASE 4.1 j/r software. The dimension and geometry were taken from the blueprint and physical measurement. All pictures shown in this report are rendered from EASE and/or EASERA.

·         Auralization

Stereo impulse responses were taken using Zoom H4 portable recorder, JBL LSR 28P speaker and 14s continuous swept sine wave (20Hz-20KHz). De-convolution and convolution process were done using CATT GratisVolver v1.2a. Recording was made on 48KHz sample rate and 16bit.

 

III.             Stage Analysis

Two stage locations are shown below (red square; ±400ft²):

Stage #1:

→north

Stage #2:

→north

Stage #1 is located between the installed sound system (speakers are also shown on the picture as small triangles). The location of this stage is ‘forced’ when a sound system is needed especially for jazz performances. Stage #2 is an alternative stage often used when the sound system is not needed. The stage #2 picture also contains the new speakers’ location, discussed later on this report.

 

Auralization

Please find the following wave files (48KHz 16bit) in the enclosed CD:

·         Dry-Rough-Track 48-16.wav

·         New – 5deg 35feet.wav

·         New – straight 25feet.wav

·         New – straight 45feet.wav

·         Old – 30deg 26feet.wav

·         Old – 65deg 23feet.wav

·         Old – straight 25feet.wav

The wave files are auralization from stereo impulse responses taken in East Hall. The position of the speaker that generates the swept sine wave is located in the middle of the stage position. The files started with ‘Old’ are auralized with the stage #1 and files started with ‘New’ are auralized with the stage #2. This auralization is meant to show the stage frequency response without any electronic supports. Please notice the change of the low mid frequency response.

The frequency response of the stage positions is shown on the next page (each line is an average from three stereo impulse measurements for each stage location). At glance, there is a comb filtering. This phenomenon happens due to the floor bounce of the swept sine wave signal. Red line refers to the stage #2 response and blue line refers to the stage #1 response.

 

  

            Analysis

Comparing the red line (stage #2 response) and blue line (stage #1 response), a stronger coloration (comb filtering) happens from the stage #1 position. A dip occurs around 590Hz and a sharp peak occurs around 190Hz. This response is auralized on the wave files (New – straight 25feet.wav and Old – straight 25feet.wav will show this) as a muddy/boxy sounding quality due to the small room resonance behind the stage.

The stage #2 response show a little more uniform response with a slight boost on frequency below 1kHz. This is caused due to the back wall (north wall) of the stage that gives more support to the low frequency. The auralization files of this stage location (referring to the files started with “New”) have a nicer and uncolored sound reproduction. Jazz performances should be reproduced with least coloration as possible, especially on the low (and low mid) frequency range to give a clearer/more defined performance.

This simple analysis shows that stage #2 has a better acoustical support for jazz performances. Some acoustical considerations are also discussed later in the report.

IV.             Installed Sound System Design Analysis

Two pairs of Renkus Heinz CT328 speakers are installed on the Hall. At glance, the low position and the aiming angle of the speakers show that the sound system performances can be improved. This chapter is focused to analyze the installed sound system performance considering each speaker has a 90 deg horizontal and 40 deg vertical dispersion. The speakers are hung 8.5ft above the finished floor and aimed ±3 deg down from the horizontal plane.

EASE Area Mapping

Direct Sound Coverage (left) and Direct to Reverberant Ratio (right)

500Hz

The stage has too much direct coverage and a little less coverage at the center line where the mixing position is usually located.

 

 

 

 

1000Hz

A good coverage is not achieved at the center line of the stage where mixing position is usually located. Direct coverage on the stage is still considered too strong.

4000Hz

Strong coverage is still considered very high at the front of the stage where the soloist or the singer is usually located. Lobing can be noticed and uneven coverage occurs in the audience area at 4000Hz.

 

 

Clarity 50 (left) and Articulation Loss (right)

500Hz

                  1000Hz

                  2000Hz

Six figures above show that the audience area does not receive a uniform clarity and articulation loss. Being used as a speech reinforcement system, the effective audience area is very limited in this set up.

Note:

·         Direct coverage and direct to reverberant ratio are shown for 500Hz, 1000Hz and 4000Hz.

·         C50 and articulation loss are shown for 500Hz, 1000Hz and 2000Hz.

·         Direct coverage legend:

Pink                        → above 100dB

Red                         → 94 – 100dB

Cyan                      → below 94dB

·         Direct to reverberant ratio legend:

Blue                        → above 0

Dark grey              → -3 – 0

Light grey              → less than -3

·         Clarity 50 legend:

Dark Blue              → 4 – 7dB

Light Blue             → 1 – 3dB

Cyan                      → below 1dB

·         Articulation loss legend:

Bright green          → 5 – 8%

Dark green            → 9 – 12%

Cyan and blue     → 13% and above

 

PAG-NAG Calculation

·         Speaker to microphone distance (D₁)             →  15ft

·         Farthest listener distance (D₂)                         → 50ft

·         Source to microphone distance (Ds)               → 0.5ft

·         Equivalent acoustic distance (EAD)               → 10ft

·         Source to farthest listener distance (Do)        → 40ft

·         Number of open microphones                         → 2

·         Feedback stability margin                               → 6dB

Note: the number of the open microphones is taken to be two, assuming that at least two microphones are being used to reinforce the soloist or singer and acoustic piano.

NAG =  = 12

PAG =  = 18.6

PAG – NAG = +6.6

Analysis

·         The PAG minus NAG value does not show a negative value, thus this system works fine in most cases. The problem in the installed sound system is mostly caused by the placement and the aiming.

·         The loudspeakers are hung way too low. This results in strong stage area direct sound coverage although the speakers are not aimed downward. From the EASE area mapping, direct to reverberant ratio is considered much higher on the stage compared to the audience area. Although the installed sound system is not being used to amplify every microphones placed for a jazz band, the sound from the speakers will bleed/leak to the recording made in the studio especially for soloist. This condition will degrade the recording quality a lot since jazz has a strong emphasize on solo section/part.

If a singer is introduced in a jazz band, the signal bleed/leak from the loudspeaker will color the lead singer’s voice. When background writing is emphasized on a top of a singer or soloist, it is going to be very hard to get a good signal on the singer/soloist and may make the system instable (feedback).

 

 

 

·         The effective audience area is basically shown below in the ellipse area:

The predicted effective audience area that has a good coverage from the loudspeakers is ±900ft².

·         A good (high) value of clarity 50ms should be obtained to give a better intimate feeling between the performers and audience. Looking at the EASE mapping analysis, good values of C50 is achieved mostly on the stage area (where the green areas are brighter), although the value is not translated bad on audience area, a better placement can be considered. This will be discussed and compared later in the report.

 

V.                New Sound System Design

This chapter is focused to improve the installed sound system performance and resolve few sound reproduction quality problems by considering another loudspeaker’s placement. Following the stage analysis chapter that concludes if the location of stage #2 gives better acoustical support, a new loudspeaker placement is shown on the stage #2 sketch on page 7. The new placement of the loudspeakers will be in the front of the stage area and 16.5ft above the finished floor. The new aiming angle is taken at 29 deg downward from the horizontal plane and going inward the room 29 deg for each speaker. Only two CT328 loudspeakers are used in this new placement. Approaches and considerations are discussed later in this chapter.

EASE Area Mapping

Direct Sound Coverage (left) and Direct to Reverberant Ratio (right)

500Hz

The CT328 loudspeakers almost omni-directionally radiate 500Hz and below. The stage is not covered by the direct sound and the lobing due to the placement is not translated badly in the audience area.

 

1000Hz

A uniform coverage of 1000Hz is achieved with this new design and the stage area is considered safe from the direct sound radiation of the loudspeakers.

4000Hz

Although less radiation on 4000Hz and higher can be noticed in the front audience area, sufficient brightness will still be achieved from the band/performers direct sound. Total coverage including reflections and sound source will be discussed later in the report.

 

 

Clarity 50 (left) and Articulation Loss (right)

500Hz

1000Hz

2000Hz

Comparing six figures on page 18 and page 12, the new stage location and loudspeaker placement will give a larger effective audience area. C50 and %ALcons values are uniformly achieved in the new audience area.

Note:

·         Direct coverage and direct to reverberant ratio are shown for 500Hz, 1000Hz and 4000Hz.

·         C50 and articulation loss are shown for 500Hz, 1000Hz and 2000Hz.

·         Direct coverage legend:

Pink                        → above 100dB

Red                         → 94 – 100dB

Cyan                      → below 94dB

·         Direct to reverberant ratio legend:

Blue                        → above 0

Dark grey              → -3 – 0

Light grey              → less than -3

·         Clarity 50 legend:

Dark Blue              → 4 – 7dB

Light Blue             → 1 – 3dB

Cyan                      → below 1dB

·         Articulation loss legend:

Bright green          → 5 – 8%

Dark green            → 9 – 12%

Cyan and blue     → 13% and above

 

PAG-NAG Calculation

·         Speaker to microphone distance (D₁)             →  20ft

·         Farthest listener distance (D₂)                         → 45ft

·         Source to microphone distance (Ds)               → 0.5ft

·         Equivalent acoustic distance (EAD)               → 10ft

·         Source to farthest listener distance (Do)        → 45ft

·         Number of open microphones                         → 2

·         Feedback stability margin                               → 6dB

Note: the number of the open microphones is taken to be two, assuming that at least two microphones are being used to reinforce the soloist or singer and acoustic piano.

 

NAG =  = 13

PAG = = 23

PAG – NAG = +10

Approaches, Consideration and Analysis

·         The original installed loudspeakers’ placement is too low and behind the stage area. This condition creates a lot of problems which degrades the quality of the sound reproduction and recording. The new loudspeakers’ placement is taken to be higher above the finished floor and placed at the front of the stage. Less sound leakage/bleed is expected due to the less direct coverage on the stage from the loudspeakers and the quality of the recording is expected to increase.

·         Higher PAG minus NAG value is achieved, thus the new placement will give more stability.

·         EASE area mapping shows that a more uniform and higher 50ms clarity value is obtained in the audience area compared to the original placement of the installed loudspeakers. This condition will give a clearer/more defined sound quality especially for jazz performances. Articulation loss on the audience area is considered low in the most of the audience area.

·         The effective audience area that is covered by the speaker efficiently and effectively is larger and shown circled on the next page.

The area circled above is predicted ±1070ft². Compared to the effective audience area covered by the original sound system installation, the new effective audience area can seat more people (more than 100 additional people).

·         The new effective audience area also shows that most of the audience sits in the front of the stage. This will give a better stereo sense of the stage. Comparing to the figure at page 15, the audience is spread across the width of the hall and stage. The sense of stereo sound reproduction is less for the audience sitting far from the center line. The stage #2 location with the new loudspeakers’ placement will give better stage’s soundscape and unity since most of the audience sits in around the center line of the stage.

·         EASE area mapping on page 16-17 show that the front of the audience area is not covered well by the direct sound of the loudspeakers especially for higher frequencies. The aiming angle of the speakers is taken to the back area of the audience to minimize over-brightness sound quality on the front due to the direct sound from the band itself.

VI.             Additional Acoustical Recommendations

Based on the stage #2 location and the new placement of the loudspeakers, few acoustical recommendations can be added to improve the quality of the music/sound reproduction. This can improve the sound recording quality as well. Several considerations are listed below.

·         The draperies are usually extended on all of the walls during a performance. The north wall draperies (behind the stage #2 location) can be opened to reinforce low energy and probably give a little brightness sensation to the hall. The draperies on the south wall should be closed to avoid a long delayed sound due to the back wall reflection from the loudspeakers.

·         Since the audience area is not sloped, a slight rise to the stage can be done to minimize the high frequency absorption due to the audience grazing angle.

·         Diffusers, such as QRD or BAD panels can be placed on the north wall (behind the stage #2 location) to help performers listening to the others. BAD panels might be preferred since they do not have a significant depth whenever the draperies need to be extended.

·         If a permanent installation of diffusers cannot be done due to aesthetic reasons, portable acoustical shells can be considered. Acoustical shells that contain diffusers and reflectors can improve the sound distribution at the stage and also redirect more energy to the audience.

 

 

·         Acoustical panels (1-2” thick) can be placed on the wall/ceiling next to the new loudspeaker placement. This will reduce the comb filtering. The picture shown below can give a rough estimate on where the acoustical panels should be placed (shown in red mark).

 

VII.          Conclusion

A stage and installed sound system analysis have been completed. A new loudspeaker placement has been developed. Acoustically, stage #2 position next the north wall gives better quality on the sound/music reproduction. The new stage position also provides a larger effective audience area.

Evaluating figures on pages 16-17, some of the areas behind the columns are not covered directly by the loudspeakers. Those ineffective audience areas (auditory wise and visualization wise) can be used for places where people want to talk just like in a nightclub or to store instrument cases/other stage equipment. Those ineffective audience areas are shown on the figure below, circled.

EASE Area Mapping

The pictures shown on the following pages are rendered to show the total coverage of the new loudspeaker placement including the reflections and the sound source on the stage. Representations on how the audience will hear the sound system are better shown with the figures on the following pages.

 

125Hz

250Hz

500Hz

1000Hz

2000Hz

4000Hz

8000Hz

The new position of the loudspeakers shows a more uniform sound coverage on the audience area and less direct sound coverage at the stage. Comparing the direct to the reverberant ratio (may not be very clear on the pictures) of the original and the new placement (page 10-11 and page 16-17), higher values of D/R ratio are obtained on the audience area with the new installed sound system due to the smaller number of the loudspeakers used.

 

Comparison

Original Stage/Loudspeaker Placement	New Stage/Loudspeaker Placement
Acoustically, the small room behind the stage #1 position colors the low-mid frequency range	Acoustically, the sound reproduction without any electronic reinforcement is flatter and better for jazz performances
Effective audience area ≈ 900 ft2	Effective audience area ≈ 1070 ft2
Direct coverage from the loudspeakers are not that uniform in the audience areas	Direct coverage from the speakers spread more uniformly in audience areas
D/R ratio is higher around the stage and more direct sound from the loudspeakers are covering the stage	D/R is more uniform in audience areas and the values are slightly higher due to the lesser amount of the speakers used
C50 and %ALcons values are not very well distributed in the audience area	C50 and %ALcons values are more uniform throughout the audience area

 

 

VIII.       References

 

G. M. Ballou. Handbook for Sound Engineers, 3^rd edition. Boston: Focal Press, 2002.

P. Brown, “Acoustic Measurements Without Computers,” Syn-Aud-Con Workshop. Dallas: February 2007.

D. Davis and C. Davis, Jr. Sound System Engineering Second Edition, Indiana: Howard W. Sams, 1987.

M. Long, Architectural Acoustics. Academic Press, 2005

 

 

IX.             Documentation