| Part 2
- How the Sweetspot was Found in One Church
Blake A. Engel, All Church Sound
In early Fall of 2002, I accompanied Joe De Buglio on a trip
to a Catholic church in Daly City, California. Located about 30
miles south of San Francisco, the church was built in the early
1970’s while initial construction of nearby (within 1,000
feet of the church!) freeway was in progress.
Once we had completed the acoustical measurements of the room,
we took many hours to take further measurements, testing for the
sweetspot. The test system being used was a MLSSA running the
latest software (V10.x). The Bruel and Kjar test microphone has
been calibrated and is within specifications which will stand
up in a court of law. This microphone was supported by a typical
3-leg tripod microphone stand. The test speaker used was a self-powered
dual-concentric variety mounted to the top of a 11-foot (minimum
height) tripod speaker stand. Prior to taking any measurements
in the room, a direct-sound test was made to ensure the test system
was operating correctly. (This type of test is performed by having
the test mic within 3 to 5 feet of the test speaker and taking
a measurement. The resulting frequency response is then compared
to a previously saved file (of the same test) used for comparison.
If the results are the same, the system is working properly. If
the results are different, some portion of the test system isn’t
working properly. This step is critical to ensure the test system
isn’t flawed.)
We initially marked the present location of the speakers in the
room using a small square of cardboard placed on the floor. We
had previously determined the speakers were too far forward, so
we then proceeded to mark locations on the floor behind this point.
Using more markers on the floor, we marked a total of 11 locations
to test. These points were in a straight line (front to back)
and were spaced randomly (some were a foot apart, others were
less than or more than, from 8 inch spacing up to 24 inch spacing).
This was done to ensure we weren’t supporting room modes
based on equally spaced test locations which may skew the results.
In determining where to place the test microphone, we used more
square scraps of cardboard. Standing at the pulpit location, Joe
“flung” many pieces out into the congregational seating
area. These markers landed at many various locations, both near
and far, some off to the side and others more toward the center
of the room. We then chose 4 locations which were a basic representation
of the randomly tossed markers.
During each test, the test speaker was aimed at the test microphone.
This was done to simulate a cluster of speakers which would be
aimed properly. Keep in mind, the test speaker used is much more
omnidirectional than a typical speaker used for sound reinforcement.
While measurements were made, all interior lights were off, all
doors and windows to the outside were closed, and the HVAC system
was turned off. Before each measurement was made, we ensured there
wasn’t an excess of noise from the nearby roads or airways
which could skew the results.
We began with mic location 1 and took a measurement with the
test speaker in each of the 11 locations we had marked on the
floor. Each time a measurement was taken, MLSSA was used to calculate
the %Alcons (intelligibility). This number was simply written
down on a sheet of paper (indicating mic position and the %Alcons
measured). This test was repeated with mic position 2, 3, and
4. The results of the measurements are show in the table below.
test speaker
location |
mic
position
1 |
mic
position
2 |
mic
position
3 |
mic
position
4 |
| 1 |
9.3% |
9.5% |
8.7% |
6.5% |
| 2 |
9.1% |
9.6% |
8.5% |
6.8% |
| 3 |
9.4% |
9.8% |
8.4% |
6.7% |
| 4 |
9.5% |
9.6% |
8.4% |
6.2% |
| 5 |
9.4% |
9.5% |
7.9% |
5.9% |
| 6 |
8.9% |
9.2% |
8.1% |
6.1% |
| 7 |
8.7% |
10.1% |
8.0% |
5.8% |
| 8 |
9.5% |
10.7% |
8.7% |
6.3% |
| 9 |
9.7% |
10.8% |
9.1% |
6.3% |
| 10 |
9.3% |
10.7% |
8.7% |
7.1% |
| 11 |
9.9% |
10.5% |
9.2% |
6.5% |
The chart shows that when the test speaker was in position 5,
6, or 7, intelligibility was the best. You may have noticed that
microphone positions were only done on one-half of the room. This
is because the room is symmetrical and what was measured on one
side would be the same as that measured on the other side (in
the same mirrored location).
This test clearly shows an increase in intelligibility when the
test speaker is in the sweetspot area rather than too far ahead
or too far behind. The increase in this church was about a half
percent. If this is the case, what limits us from using the sweetspot
(rather than some other location) to gain more intelligibility
for no extra cost? Some may argue that mounting location for the
speaker(s) is more convienient if they’re placed outside
the sweetspot. This may be true in some cases, but not all.
This church had a large cupola in the center which was roughly
16 feet wide and 20 feet tall. Thankfully, the walls were angled
which meant it wasn’t as bad as it could be. The people
standing under this structure do hear a noticeable bass boost,
however.
Even if you’re unable to locate your speaker(s) in the
sweetspot, you should be aware of the location to avoid putting
the pulpit there or any other open microphone. This situation
would cause excess trouble with feedback. Avoiding this location
for open mics is using your knowledge of the sweetspot to your
advantage, making you look all the more wise!
A Few Final Comments
In this example, the test speaker used had a very low Q. That
is, the speaker is not very directional. With a proper sound reinforcement
speaker that included a horn of proper size, the results are much
better. (In one church Joe tested, he measured higher intelligibility
with his low-Q test speaker in the sweetspot over the existing
horn-loaded speaker system that was out of the sweetspot. Thus,
if you can get better intelligibility from a low-Q device that’s
in the sweetspot compared to the higher-Q device out of the sweetspot,
imagine how good the intelligibility would be if the higher-Q
device was in the sweetspot!)
Click here for Part 3 |
