The
question of whether or not internally bracing a speaker cabinet has come
up for discussion many times. This recent time prompted the
following experiment. Unfortunately, an accelerometer had to be
substituted with a microphone
The initial peak in both these curves is 33 dB below the initial peak measured at the front of the diaphragm at a distance of about 0.030". Both measurements were performed with the diaphragm moving inward in order to avoid the diaphragm's striking the microphone. The speaker was connected to a 9v battery through a SPST switch and the trace captured with a Pico Scope, a digital oscilloscope connected to a computer. Similar traces were obtained on a Tektronix 465 analogue scope but taking a photo was impossible due to the lack of storage capability. The mic output was fed through a vacuum tube ART Tube MP microphone amplifier to boost the signal from the few millivolts output of the mic which was too low for the Pico Scope's lowest 50mV/division scale. In short, it seems that for smaller cabinets of less that 3 cubic feet, internal bracing is not really necessary. This will probably change as the cabinet walls become larger. Considering the 25.21 dB difference between the speaker output and that from the rear, a power factor of 332:1, the output from panel vibration won't be heard. What many probably hear is vibration from weak or loose joints in the cabinet. The sound introduced from panel vibration would be far less than the vibrations of items in the room, like walls for one, not to mention picture frames and just about anything else that isn't nailed down, especially when playing at very loud levels. If a cabinet is well built using a good grade of 0.75" birch plywood or MDF and cleats used internally, the need for bracing seems to be overkill. However, if one is constructing a bass cabinet and doesn't mind the extra work and if it makes that person feel better, then do it. This experiment was performed to show the difference between a braced and un-braced cabinet, not to discourage anyone from using braces.
|
The
cabinet innards showing the braces. Originally, the front was
glued to the front to rear brace but has been cut for some unknown
reason. The square extension was later added and protrudes about 0.030"
into the front. This allows the front to crown when the screws are
tightened, ensuring a very stiff front and rear.
The side to side brace is also about 0.030" wider then the internal width and had to be tapped into position. The internal volume is 3.0 ft^3 and was brick filled to reduce the internal volume to 2.44 ft^3, the optimum for the Sansui W200 speaker. This gives an optimum Q of 0.707 for this sealed enclosure which means the diaphragm is optimally damped. The innards were heavily stuffed with 3 pieces of fiberglass measuring 4" thick and cut to fit the inside dimensions. |
This is
a Shure microphone with the shells removed. This allows the mic
diaphragm to be placed 0.030" from the rear surface
otherwise, the closest would have been about 0.4375"
(7/16ths inch)
The shim was removed for the test. An accelerometer is on order from Newark Electronics which will give much better results. |
A front
view of the cabinet used for the test.
As can be seen, the vent is covered. The barrier strip on the left was originally used for testing a Jensen F12JN field coil speaker. The two upper screws were to connect the field coil and the lower ones, the voice coil. This is one of a pair of cabinets that were placed in storage due to a design that was not completed. |
Pulse
from the rear of the un-braced cabinet in the geocenter, the point of
maximum vibration.
The vertical scale is 100 mV/division; horizontal scale is 20 mS/division The pulse width is 60 mS (between the green lines) |
|
|
Same
position and scales as above but cabinet is braced in two directions,
side to side and front to rear.
The two traces are very similar with exception that this pulse width is 53mS, 7 mS less than the undamped trace.
|
|
|
Back to the loudspeaker main page